Cover Page

The handle http://hdl.handle.net/1887/43970 holds various files of this Leiden University

dissertation.

Author: Power, R.C.F.

Title: Evaluating the dietary micro-remain record in dental calculus and its application in
deciphering hominin diets in Palaeolithic Eurasia
Issue Date: 2016-11-01
Evaluating the dietary microremain record in dental
calculus and its application in deciphering hominin
diets in Palaeolithic Eurasia
Proefschrift

ter verkrijging van

de graad van Doctor aan de Universiteit Leiden,

op gezag van Rector Magnificus prof. mr. C.J. J. M. Stolker,

volgens besluit van het College voor Promoties

te verdedigen op dinsdag 1 november 2016

klokke 16.15 uur

door

Robert Charles Fergal Power

Geboren te Waterford, Ierland

in 1989
Promotor: Prof. dr. Jean-Jacques Hublin (Universiteit Leiden and the Max-Planck-
Gesellschaft)

Co-Promotor: Dr. Amanda G. Henry (Max-Planck-Gesellschaft)

Co-Promotor: Dr. Domingo C. Salazar-García (University of Cape Town, and the
Max-Planck-Gesellschaft)

Promotie commissie:

Prof. dr. Wil Roebroeks (Universiteit Leiden)

Dr. Alexander Verpoorte (Universiteit Leiden)
Prof. dr. Annelou van Gijn (Universiteit Leiden)
Dr. Geeske Langejans (Universiteit Leiden and the University of Johannesberg)
Prof. dr. Marco Madella (Universitat Pompeu Fabra)
Copyright © Robert C. F. Power 2016
All Rights Reserved

Cover design by Mariska Carvalho

This research has been made possible through a Max Planck Independent Research
Group grant of the Max Planck Society, obtained by Dr. Amanda G. Henry. The
research was supervised by Dr. Amanda G. Henry and Dr. Domingo C. Salazar-
García.
Contents

Preface.................................................................................................................................................... 1
Introduction: the evolutionary context of Neanderthal dietary ecology ..................................... 3

Assessing use and suitability of scanning electron microscopy in the analysis of

microremains in dental calculus ...................................................................................................... 47

Dental calculus evidence of Taï Forest Chimpanzee plant consumption and life history
transitions ............................................................................................................................................ 67

Dental calculus indicates widespread plant use within the Neanderthal dietary niche ......... 91

Discussion: A pathway for reconstructing Neanderthal Dietary Ecology............................... 119

References ......................................................................................................................................... 133
Appendixes ....................................................................................................................................... 165

Summary ........................................................................................................................................... 255

Samenvatting .................................................................................................................................... 257

Acknowledgements ......................................................................................................................... 259
Contributions .................................................................................................................................... 263
Curriculum vitae .............................................................................................................................. 265

v
List of Figures

Figure 1: The largest known range of Neanderthals ....................................................................... 7

Figure 2: Conceptional illustration of a diet breadth model ........................................................ 12

Figure 3: Carbon wt % of starch, sugars and sugars produced by hydrolysis .......................... 56

Figure 4: Carbon wt % comparing native starch versus hydrolysed samples .......................... 56

Figure 5: Starch grains located in-situ on dental calculus surface ............................................... 58

Figure 6: SEM image showing microremain diversity ................................................................. 59

Figure 7: SEM image showing localised damage that arises from higher primary voltage .... 59

Figure 8: A calcium oxalate prism observed with optical microscopy....................................... 62

Figure 9: Starch and phytolith morphotypes used in the identification model ........................ 78

Figure 10: Microremains recovered in dental calculus samples.................................................. 80

Figure 11: Microremain assemblages recovered in calculus ........................................................ 81

Figure 12: Plant genera represented by microremains and Chimpanzees diet......................... 83

Figure 13: Mixed poisson regression model predicted values .................................................... 85

Figure 14: Abundence of Coula nut starches with chimpanzee age at death............................. 87

Figure 15: Map of western Eurasia with the studied sites indicated .......................................... 96

Figure 16: Miroremains from Neanderthal calculus, fauna calculus and controls ................. 109

Figure 17: Mosaic of microremains and modern reference plant matter ................................. 112

Figure 18: A Menhinick’s index of types of starch and phytolith and climate ....................... 114

Appendix figure 1: Starches per mg in each chimpanzee sample and year of death............. 175

Appendix figure 2: Chimpanzee plant foods, ranked by minutes consumed ........................ 176

Appendix figure 3: Total numbers of starch and phytoliths in each Neanderthals site ........ 251

vi
List of Tables

Table 1: Energy yields of various food classes consumed by recent foragers ........................... 16

Table 2: Neanderthal sites with evidence of macrobotanical plant remains ............................. 28

Table 3: Neanderthal remains with published stable isotopic values ........................................ 37

Table 4: Calculus samples analysed using SEM-EDX and OM ................................................... 52

Table 5: List of reference samples analysed using EDX ............................................................... 53

Table 6: Recovered microremains using both microscopy approaches. .................................... 61

Table 7: Plant genera selected from reference collection for the identification model............. 71

Table 8: All chimpanzee dental calculus samples analysed......................................................... 73

Table 9: Neanderthal dental calculus samples ............................................................................. 102

Table 10: Palaeoenvironmental simulations used to predict seasonal temperature .............. 105

Table 11: Palaeoenvironment reconstructions for each specimen used in this study ............ 106

Appendix table 1: Elemental composition of standards ............................................................. 165

Appendix table 2: Elemental composition of degraded and native starch .............................. 166

Appendix table 3: Elemental composition of calculus and microremains in calculus ........... 168

Appendix table 4: Inventory of analysed plants and fungi ....................................................... 177

Appendix table 5: Additional details of Chimpanzee calculus samples .................................. 181

Appendix table 6: Metrics of reference phytoliths and starches ............................................... 182

Appendix table 7: Microremain variables used for identification model ................................ 202

Appendix table 8: Random forest phytolith identification model ............................................ 202

Appendix table 9: Random forest starch identification model .................................................. 203

Appendix table 10: All recovered microremains in each dental calculus sample .................. 204

Appendix table 11: Counts of identified genera in Taï Chimpanzee calculus samples ......... 208

Appendix table 12: Measurements of phytoliths and starch from calculus ............................ 208

Appendix table 13: Coefficients of statistical models ................................................................. 237

Appendix table 14: Variable importance in phytolith and starch random forest ................... 238

Appendix table 15: Total recovered microremains from Vindija .............................................. 243

Appendix table 16: Total recovered microremains from Grotta Guattari ............................... 246

Appendix table 17: Total recovered microremains from Grotta Fossellone ............................ 248
vii
Appendix table 18: Total microremains from Sima de las Palomas del Cabezo Gordo ........ 249

Appendix table 19: Total microremains from Kalamakia .......................................................... 250

Appendix table 20: Coefficients of statistical models ................................................................. 252

Appendix table 21: Western Eurasian starch-rich economic plants ......................................... 252

viii
 1

Preface

Since William King first described Neanderthals as a distinct species in 1864,

this hominin has provoked discussion, insight, and debate. This event in taxonomic
history was a breakthrough in understanding human origins. Since this milestone,
we have learned much more about Neanderthals and their relationship to ourselves.
The application of ecological perspectives to this topic has addressed more
comprehensively than ever before the subsistence, behavioural ability, and the
ultimate fate of this extinct hominin. This doctoral thesis aims to continue the spirit
of these advances by examining how frequently plants featured in Neanderthal
dietary regimes and assessing what this means for the apparent distinctiveness of
Neanderthal diets. It does this by developing dental calculus analysis as an
archaeobotanical technique and exploring how it can reveal Neanderthal plant use.
The scope of this thesis encompasses the development of high-resolution approaches
for reconstructing food choice using dental calculus, and the use of these advances to
revaluate Neanderthal diets in the context of Pleistocene ecological conditions. This
thesis is comprised of six chapters, of which the middle three are stand-alone papers,
two of which are published and one submitted for publication:

a) Introduction: The evolutionary context of Neanderthal dietary ecology

(Chapter 2).

b) Assessing use and suitability of scanning electron microscopy in the analysis

of microremains in dental calculus (Chapter 3). Published as Power RC et al., J
Archaeol Sci 49:160–169. 2014.

c) Dental calculus evidence of Taï Forest Chimpanzee plant consumption and life
history transitions (Chapter 4). Published as Power RC et al., Sci. Rep 5. 2015.

d) Dental calculus indicates widespread plant use within the Neanderthal dietary
niche (Chapter 5). Submitted as Power RC et al. to the J Hum Evol.

e) Discussion: a pathway for reconstructing Neanderthal dietary ecology

(Chapter 6).

1
 My thesis initially contextualises the significance of the Neanderthal diet
within the broader evolution of hominin diets. The importance of Neanderthal diets
for understanding this Pleistocene hominin and explaining its fate are examined. It
also discusses how dental calculus can assist dietary studies of this hominin. The
first paper examines the shortfalls of conventional analytical dental calculus
approaches, and develops a high–resolution workflow for optimising extractable
dietary data from dental calculus. The second paper explores the representativeness
of the dental calculus dietary record. It quantifies its resolution with dental calculus
samples from chimpanzees with a documented dietary history. The third and last
paper examines if the Neanderthal dental calculus dietary record from a variety of
sites situated across their range in time and space suggest flexible or rigid diets. I use
these results to place their diet within the current knowledge of hominin plant use,
hominin dietary breadth, and the evolution of hominin diets.

2
 2

Introduction: the evolutionary context of Neanderthal

dietary ecology

2.1 The importance of understanding Palaeolithic diets

Energy provisioning is central to mammalian ecology. Understanding

mammalian life requires examining its acquisition of energy. Dietary details, such as
nutritional requirements, use of different environments, and trophic level may
define a taxon (Hutchinson, 1959; Armstrong and McGehee, 1980). The importance
of diet is no less true for our own lineage (Winterhalder and Smith, 2000). Dietary
strategies are relevant for understanding the most basic and the most specialised
level of organisation of human societies. Exposure to new environments and climatic
cycles forced major subsistence changes because these presented energetic
challenges that restricted, expanded, or otherwise shaped past populations. Scholars
recognise that reconstructing ancient dietary niches is of central importance to early
hominin palaeobiology. Many authors have suggested that niche switching is linked
to landmark changes in hominin evolutionary history (O’Connell et al., 1999;
Wrangham, 2000). This is evident from the emergence of distinct morphological and
behavioural changes.

Among early Pleistocene hominins, for example, differences in the apparatus

of mastication distinguish robust and gracile forms. Enormous, thickly enamelled
teeth and robust mandibles that supported powerful mastication ability separate
robust Paranthropus boisei and Paranthropus robustus from the less-developed forms
including Australopithecus afarensis and Australopithecus africanus. Chipping of the
enamel from the robust species is suggestive of the consumption of hard foods
(Constantino et al., 2010), although microwear fails to support this view (Ungar et
al., 2008). The ability to consume extremely hard and tough foods may have been an
adaptation for processing fallback foods during seasonal bouts of food scarcity (Lee-
Thorp et al., 1994; Ungar et al., 2008; Constantino et al., 2010). Stable isotopic
evidence has told a different story on why this specialised robust anatomy evolved.
Paranthropus robustus isotopic values suggest this hominin was reliant on C4 grasses
and sedges, in contrast to those of nearly all other hominins, which show more
reliance on C3 resources. This implies that the robust craniodental features are an

3
adaption to “repetitive loading” from consuming large quantities of low-quality
vegetation rather than hard objects (Cerling et al., 2011). These craniodental traits
represent differences in adaptive capabilities from more gracile forms even if their
dietary niches were composed of similarly tough foods (Laden and Wrangham,
2005; Ungar et al., 2008).

Just as early Pleistocene robust hominins morphologically adapted to a

dietary niche, other hominins have evolved morphologically in many other ways
because of exploiting new dietary regimes. A dietary change is intertwined with the
most dramatic externalisation of hominin evolution, the dramatic increased size of
the hominin brain. According to the “expensive tissue hypothesis,” the emergence of
a large and energetically costly large brain and small gut in the Homo genus arose
concomitant to an increasingly energy-dense diet (Aiello and Wheeler, 1995) . The
trade-off between a larger brain size and a smaller digestive tract demanded a higher
quality diet, which was possible through either consuming more high-energy classes
of food or through external digestion (i.e. food processing or cooking). Some of the
most recent anatomical changes in hominin history relate to diet. Present day human
dentition is the outcome of a progressive reduction in the size and number of teeth
since the first emergence of Homo. The process of diminishing tooth size accelerated
in periods of dietary change such as the transition to agriculture (Loring Brace et al.,
1987). A decrease in the loads exerted on teeth due to changing properties of diet
explains this process, whether it is a phenotype relaxation, or selection for a
reduction in size and number of teeth.

Other subsistence patterns that have developed recently in hominin history

have also left their mark on parts of our biology. For example, the emergence of
agriculture has instigated several distinctive changes that remain with us. The AMY1
gene, which is responsible for producing salivary amylase, the enzyme that breaks
down starch in the mouth by hydrolysing it into more useable sugars, is one such
example. Present day humans have several copies of this gene (six is the average),
and the more copies an individual has, the more amylase is expressed in their saliva
(Bank et al., 1992). AMY1 copy number variation follows a gradient, such that more
copies are present in populations with historically heavy intake of starch, and few
copies occur in populations with a low intake of starch (Squires, 1953; Perry et al.,
2007; Carpenter et al., 2015; Hardy et al., 2015a). This indicates that some copy
number variation evolved recently (during the last 10,000 years) in response to the
proliferation of starch in diets from agriculture (Perry et al., 2007). The impact of the
lower copy phenotype, a reminder of our hunter-gatherer past, reverberates today as

4
a lower estimated AMY1 copy-number is linked to obesity and morbidity (Falchi et
al., 2014).

As we have seen, our morphology and genome testify to millions of years of

changing diets. Dietary adaptations are present from the earliest hominins in early
Pleistocene Africa to recent populations in the last 10,000 years. Unsurprisingly, in
the case of Neanderthals, subsistence strategies have been a preeminent focus of
research. Just as diet was likely the main trait that differentiated early robust and
gracile hominins, dietary behaviours are thought to have set Neanderthals apart
from other hominins. Neanderthal diet has been suggested as narrow, specialised
and profoundly conservative, unlike that of early modern humans, and this dietary
niche influenced their range, population history and disappearance.

Neanderthals were closely related to early modern humans, and are even
known to have interbred with them, but were distinct in anatomy, ontogeny and
techno-cultural expression (Spoor et al., 2003; Smith et al., 2007; Klein, 2009; Gunz et
al., 2010; Murray et al., 2015). It is less clear if their diet differed from that of early
modern humans. The apparent distinctiveness of Neanderthal resource use has led
researchers to link it to their displacement at the end of Middle Palaeolithic. Some
consider that Neanderthal diet was reliant on a more restricted range of animal food
staples than that of early modern humans (O’Connell, 2006; Stiner, 2013). An
inflexible subsistence pattern, due perhaps to cultural or biological factors, may also
have burdened Neanderthals with a competitive disadvantage when Upper
Palaeolithic modern peoples began to enter Eurasia. Certainly, the more
Neanderthals ascended the food chain the more prone they were to experiencing
episodes of insufficient food supply. Potentially, this might explain their small
isolated populations, their history of regional extinction events and displacement.
Unfortunately, it is unclear if their foraging strategies were as inflexible as imagined
in this scenario. Furthermore, it is unknown if their economy responded to different
ecologies or was static across their range. Understanding their plant use in particular
is fundamental because plant use has major implications for their trophic position
and the adaptability of their diets. However, because we have limited knowledge on
Neanderthal plant use we thus cannot answer these questions. To assess what diet
may reveal about this hominin we must first consider Neanderthal origins and
history.

5
 2.2 Neanderthals: phylogeography and chronology

2.2.1 Neanderthal origins

Neanderthals evolved from hominins of African origin that entered western

Eurasia at some point in the Pleistocene (Hublin, 2009). Revised genetic evidence
indicates that the lineage that gave rise to Neanderthals split from the ancestors of
early modern humans 550 to 765 ka depending on the pace of the mutation rates
(Meyer et al., 2014; Prüfer et al., 2014). A date of about 500 ka would agree with
hominin remains found in Europe at this time. Dental morphological evidence
poorly converges and it may suggest a split as early as one million years ago
(Gomez-Robles et al., 2013). Due to the breadth of these estimated time ranges, and
discrepancies between the different lines of evidence, the last common ancestor
species is contentious and the geographic setting where it evolved is unclear.
Leaving this aside, there appears to be evolutionary continuity in morphology from
the hominins found in Europe dated to 500-300 ka (Arago in France, Sima de los
Huesos in Iberia, Petralona in Greece and Mauer in Germany) to Neanderthals.
Arguably early African fossils have no such continuity to Neanderthals (Bermúdez
de Castro, 1997). Palaeoanthropologists can only reliably assign skeletal remains to
Neanderthals in the late Middle Pleistocene (230-180 ka) at European sites such as
Biache-Saint-Vaast, Fontéchevade, La Chaise Suard, and La Lazaret in France
(Churchill, 2014). Most of our knowledge about Neanderthals stems from their later
chronological range from 130-30 ka.

2.2.2 Neanderthal range

Neanderthals are known from sites throughout Eurasia, such as Forbes’

Quarry, Devil’s Tower, Zafarraya, El Sidrón in Iberia; Le Moustier, La Ferrassie,
Regourdou, Pech-de-l’Aze, Roc de Marsal and La Chapelle-aux-Saints in France;
Neanderthal in Germany; Grotta Guattari in Italy; Krapina and Vindija in Croatia
and Kůlna in the Czech Republic; Teshik-Tash in Uzbekistan; Shanidar in Iraq;
Amud, Tabun and Kebara in Israel. Although applying a species concept to
Neanderthals or any extinct hominin can be difficult and inevitably controversial,
the skeletal remains from many of these sites exhibit morphological traits that typify
Neanderthals. These sites suggest Neanderthals lived in much of western and
central Eurasia. They occupied as far north as the German Coast while in the south
their ranged stretched to the Mediterranean rim, the Levant and parts of Iraq (Fig. 1).

6
Their east to west range stretched from Atlantic Iberia in the west and central Siberia
in the east. Skeletal remains show conclusively that they lived as far east as the Altai
Mountains in central Asia (Krause et al., 2007). In this span of western Eurasia, there
is variation and discontinuity. Occupation of northern areas varied according to
glacial cycles (Van Andel et al., 2004). This is evident in the depopulation of northern
areas in cold periods of MIS 6 and MIS 4, due to either the volatility of climate or the
harshness of the climatic conditions themselves. These regions were subsequently
recolonised in warmer phases. Archaeological and mitochondrial DNA evidence
shows that this process occurred through a process of incremental local extinctions
in northern zones on the onset of cold phases rather than Neanderthals tracking the
movement of milder climates south (Hublin, 1998; Roebroeks et al., 2011).

Fig. 1: The area shaded in blue represents the largest known range of Neanderthals based on lithic and
skeletal evidence. Krause et al., 2007 modified by Ryulong licenced under CC-BY-SA-3.0 and by author.

2.2.3 Neanderthal disappearance and its implications for dietary ecology

Although Neanderthals survive in part to this day as archaic DNA in the

contemporary human genome, Neanderthals are an extinct branch of humanity. The
manner of their disappearance has proven to be difficult to resolve. They roamed
Eurasia well into the warm MIS 3 Phase (the interplenniglacial), but how late they
survived and if their disappearance is a result of the spread of Upper Palaeolithic
early modern people is hotly debated (Finlayson, 2008; Pinhasi et al., 2011; Higham
et al., 2014). Late Neanderthals may have survived in far-flung pockets of their
range, including in Southern Iberia where there are few Aurignacian remains

7
(Finlayson et al., 2006), and in the Caucasus Mountains (Ovchinnikov et al., 2000). In
addition to these postulated refugia, a northern refuge near the Arctic Circle at
Byzovaya has been suggested based on Mousterian tools dating to 34-31 ka cal BP
(Slimak et al., 2011). This is an exceptionally northern site, outside of the
conventional views of their range and it fits poorly with available data (Zwyns et al.,
2012). Excavators found no hominin remains, meaning they were unable to clarify if
this is a Late Neanderthal occupation. The evidence verifies that Neanderthals were
gone across their range by 33 ka cal BP, but they could well have disappeared
considerably earlier as dates this late are rare (Galván Santos et al., 2006; Wood et al.,
2013a). Many argue that they survived no later than 40 ka cal BP (Pinhasi et al., 2011;
Wood et al., 2013b; Higham et al., 2014; Hublin, 2015). One reason why this problem
is difficult to resolve is that the period is at the temporal limit of the applicability of
radiocarbon dating (Higham, 2011).

Both changes in stone tool technologies and variation within technocomplexes

have influenced our interpretation of the disappearance of Neanderthals.
Neanderthals developed a stone tool industry called the Mousterian Industrial
Complex, characterised by the presence of large, specially prepared cores and
specialised flakes often made using the Levallois technique (Klein, 2009). This is the
dominant technology in western Eurasia until the arrival of early modern Upper
Palaeolithic technocomplexes such as the Aurignacian. In some regions, the
archaeological layers containing Neanderthal-associated Mousterian tools are
separated from those containing Aurignacian tools by layers containing artefacts
from so-called transitional industries. One of the best-studied examples is the
Châtelperronian of central France and northern Iberia. Many aspects of the
Châtelperronian are characteristically Upper Palaeolithic, leaving some to wonder if
it was manufactured by Neanderthals (Ruebens et al., 2015). The Châtelperronian is
the only transitional industry with directly associated Neanderthal remains (Hublin
et al., 1996), strongly indicating that it was, in fact, created by Neanderthals.
Reassessment of Châtelperronian tools appears to suggest that this complex emerged
from the local Mousterian (Granger and Lévêque, 1997; Ruebens et al., 2015). This
raises the question of Neanderthals groups interacted and exchanged culture with
Upper Palaeolithic early modern humans. Resolving this issue is central to
understanding the potential intensification of plant use suggested by
Châtelperronian grindstones (See 2.5.1). If a process of acculturation occurred, it
could have influenced multiple levels of Neanderthal culture including their diets.
Debate has centred on whether the Châtelperronian appeared following contact with
early moderns. Some have argued that the Châtelperronian was manufactured by

8
Neanderthals before Upper Palaeolithic modern humans arrived in Europe and that
its stratigraphic overlap with moderns at the key Châtelperronian site of Grotte du
Renne is a product of sediment disturbances and layer remixing and cannot be
reliably interpreted (Zilhão et al., 2006). Indeed, at Grotte du Renne (Arcy-sur-Cure,
France), there is evidence of remixing in the sequence (Higham et al., 2010).
Reattempts at dating imply that Neanderthals were the makers of the
Châtelperronian industry, and the Châtelperronian Neanderthal (Saint-Césaire)
post-dated the arrival of early modern humans in western Europe (Hublin et al.,
2012). This timing suggests a cultural diffusion from modern to Neanderthal groups.

We can also discern interaction by identifying gene flow between these

hominins. Geneticists have pinpointed multiple events of introgression between
Neanderthals and early modern humans. This introgression likely occurred in the
Near East prior to the split of the ancestors of present day Europeans and Asians (86-
37 ka) (Sankararaman et al., 2012). A second event may have occurred, presumably
further east concerning the ancestors of present day Asians only (Vernot and Akey,
2015). An early modern human dated to 42-37 ka from Peştera cu Oase, Romania,
one of the oldest modern humans found in Europe, showed recent admixture with
Neanderthals, with 6–9 % of its genome from a Neanderthal ancestor. The
completeness of the archaic DNA in the Oase individual indicates that this cross
occurred four to six generations prior. The recent suggested date of this admixture
indicates that admixture probably occurred in Europe (Fu et al., 2015).

What this cultural diffusion and population admixture tells us about the
Neanderthal subsistence and its ability to adjust to new circumstances is unclear.
The presence of two or more types of hominins in Eurasia inevitably led to
overlapping territories. Both hominin groups would have sought the same high
quality resources, leading to direct competition for the optimal foods. The impact of
new hominins on Neanderthals would vary according to how numerous
Neanderthals were; a large population could buffer against a large influx of
competing hominins. Yet high-quality ancient genomes have revealed that
Neanderthal demographic history differs from that of early moderns. Neanderthal
genetic history displays a protracted history of small isolated populations and low
genetic diversity (Castellano et al., 2014; Prüfer et al., 2014). Small populations could
have left them highly vulnerable to even minor competition from early moderns.
The inevitable increase in isolation may have reduced their ability to develop
resilient subsistence patterns to cope with the arrival of early moderns.

9
 Neanderthal admixture, decline and extinction may imply that their niche
was susceptible to competition. It is easy to envisage early moderns who arrived in
Europe creating ecological imbalances that disrupted Neanderthal foraging. Given
chronological ambiguity, this is not currently detectable. On the other hand,
competition may have led to an intensification of Neanderthal subsistence. The
cultural diffusion may have led to Neanderthals adopting modern subsistence
strategies, but this elucidates little about the dietary niche Neanderthals used for
200,000 years previously. To interpret their long-lived subsistence and dietary
regime we must examine their diets using an interpretive framework loaned from
ecological theory that allows us to make predictions about their foraging behaviour.

2.3 Applying a framework for studying ancient diets

2.3.1 Human Behavioural Ecology

Knowledge of Palaeolithic human diets is not useful if we have no means of

interpreting this information. A theoretical framework that allows us to place dietary
choices in a cultural and biological context is needed. Human Behavioural Ecology
(HBE) is a useful framework for studying dietary choices and the environmental,
biological, and cultural limitations that frame those choices. Behavioural ecology
emerged in the 1960s and the 1970s amongst circles of ecological theorists seeking a
basis for studying feeding, social, and reproductive behaviours (Bird and O’Connell,
2006). Behavioural ecology spread to researchers interested in human societies, as it
allows human behaviour to be interpreted in the rubric of evolutionary ecology
theory. Human behavioural ecology posits that individuals tend to adapt to their
environment as necessary to maximise their fitness (Mulder and Schacht, 2012). On a
daily basis, foragers will consider and weigh decisions on their costs and benefits. If
behaviours diverge from this pattern, the possibility of social and cultural factors can
be investigated. Behavioural ecologists attribute behavioural diversity to the
cumulative impact of the strategies of individuals, the local ecological niches, and
the cultural transmissions of information (Smith, 2011; Mulder and Schacht, 2012).

Human behavioural ecology has been widely adopted in anthropology as it

offers a framework to generate testable hypotheses about behavioural variation. One
group of HBE theories that are popular in studies of human origins are optimal
foraging theory and diet breadth models. These models provide a powerful way to
assess forager feeding strategies. In a diet breadth framework, researchers predict

10
whether a forager will collect a potential food item that it encounters while foraging
(MacArthur and Pianka, 1966). To acquire a resource, the forager must bear both the
search costs (e.g. location, hunting and digging) and the handling costs (e.g.
preparing and processing), and will minimise these as far as possible although this
does not influence a resource’s ranking in the diet breadth model. A forager can seek
to maximise their efficiency by incorporating the most profitable food items (plant or
animal) available and ignoring lower ranked prey (Fig. 2). The inclusion of any given
item relies on its ultimate profitability to the forager rather than its abundance (Bird
and O’Connell, 2006). Researchers usually assume that foragers assign a rank to a
potential food depending on its energy yield minus the cost of food preparation, but
foods may also be ranked on other currencies, such as specific macro or
micronutrients that are more physiologically important to the individual than total
energy (Hockett and Haws, 2005). Food may even be ranked according to individual
subjective preferences. The foraging cultures of Alaskan peoples give potential real
world examples of non-energetic currencies. Alaskan foragers commonly rob
nutritious foods such as sedge corms (Cyperaceae spp.) from rodent caches and
nests, which rodents collect for winter food. They compensate the loss of their
plunder with fish to sustain the rodent over the winter (Anderson, 1939). A desire to
procure vegetable nutrients, rather than energy alone probably explains this
behaviour. A low-ranked food item may also be a prey individual that is younger
and smaller than normal for the taxon. A food type may be highly ranked, energy-
rich and abundant but fast moving, and hard to catch and hence rarely entering
diets. Improved technologies can dramatically lower handling costs (Kelly, 1995).
For example, nets and weirs majorly abate the search costs for catching river fish,
although maintaining nets and weirs is a substantial long-term cost and a constraint
on mobility. For this reason, recognising technological change is important for
deciphering foraging choices.

11
Fig. 2: Conceptional illustration of a diet breadth model that uses energy. Behavioural ecologists rank foods
on their energy (or other nutritional metric) return net of the processing costs expended to obtain energy.
Ranking in diet breadth models ignores prey abundance and prey search costs.

Given current data on handling costs, diet breadth models predict that due to
the differences in energy yield, animal foods (especially meat and fat from medium
and large ungulates) and honey are high-ranked, in contrast to the majority of plant
foods (Table 1). However, plant foods are usually a more abundant staple and their
collection is far more reliable than most types of hunting. Nevertheless, because of
their laborious collection and handling most plants provide fewer calories (Kelly,
1995; Kuhn and Stiner, 2006), and are less preferred. Ethnographic studies of food
preferences confirm this trend (Berbesque and Marlowe, 2009). Although recent
foragers prefer animal foods, plant foods are still a ubiquitous feature of their diets.
This is readily explained by the fact that the quest for food relies on reducing food
supply volatility, rather than always attempting to maximise the amount of food one
takes in (Winterhalder, 1986). Plants are usually more abundant and occur in large
patches, they are more easily and reliably located than game. This is important
because forager mobility is limited.

A lesson in the value of these diet breadth models in interpreting food choice
may be found in the classic ethnography of the Aché of eastern Paraguay (Hawkes et
al., 1982). The Aché hunt peccary, deer, and monkey, and they gather oranges, palm
hearts, palm fibre and larvae in palms as well as honey. Hunting offers unreliable

12
returns yet the Aché are predominantly hunters. Game provide about ~78 % of
dietary energy while plants provide only ~12 % (Gurven et al., 2001). Plants and fish
available to the Aché offer more abundant, more reliable, and far more readily
located foods. Aché hunting game will stop and collect honey, grubs, and the most
highly ranked plants they pass such as oranges. Yet they often ignore other plant
foods like palm fibre even though these are regularly consumed when little else is
available. The Aché provide several lessons for studying ancient hominin food
provision. Examination of energy returns reveals that hunting is typically the
optimal strategy, although unreliable, hunting offers the highest overall returns.
Although plants provide less energy (Kelly, 1995), they are crucial and ensure
survival in periods of shortage. It also is a reminder of the uneven distribution of
resources in the landscape (Hawkes et al., 1982). While this patchiness is
unaccounted for in diet breadth models it is incorporated into other theories such as
the patch choice model and the marginal value theorem. Although these theories are
difficult to apply to Palaeolithic societies, they further illustrate why plants are
collected for subsistence. The patchy nature of food distribution combined with
labour specialisation favours the common use of both high- and low-ranked foods.
This strategy is true in most societies worldwide.

Recent foragers exploit variations in the reliability and abundance of plant

and animal foods with highly efficient systems of labour specialisation, usually
according to sex. Generally, in recent forager societies females collect most plant
foods while males tend to hunt, especially when the quarry is large and dangerous
(Kelly, 1995).

This sex specialization pattern is true in almost all reported recent forager
societies, except the Agta of the Philippines (Goodman et al., 1985). One advantage
of this system is that it accommodates the difficulties that a highly mobile hunting
lifestyle brings to aspects of reproduction such as breast-feeding. The gathering of
plants and immobile animal foods like shellfish is more conductive than hunting to
the feeding demands of infants, and reduces the risk starvation because gathering
yields nutrients even when hunting fails (Brown, 1970). Moreover, hunting large
game may have been useful to males as a method of building social status by
signalling fitness to the group, this is known as “costly signalling” (Hawkes and
Bird, 2002). A division of labour between hunting and foraging thus optimises the
high-energy returns of meat and the reliability of plant collection. The risk of a diet
that eschews plant food and depends on animal food is observed in the early records
of arctic foraging peoples. Little plant nutrition was available to arctic foragers and

13
this coupled with extreme seasonality meant that stretches of hunger and even
devastating starvation were a regular feature of life (Ackerknecht, 1948; Young,
1996). Records from isolated communities in eastern Greenland show that as many
as 15 % of the population died of starvation between 1881 and 1883 (Holm, 1911).

2.3.2 The subsistence trajectory of Palaeolithic societies

Diet breadth models provide a starting point for exploring subsistence choices
of past groups. They have formed the dominant paradigm to explain Palaeolithic
subsistence. Prehistorians have invoked behavioural ecology to argue that there is a
landmark point in hominin dietary history at which foragers switched prey in a
transition termed the “Broad Spectrum Revolution” in the terminal Pleistocene
(Binford, 1968; Flannery, 1969; Zeder, 2012). Hunters of medium and large game
began to heavily rely on plants, fish, and fowl due to climatic and demographic
pressures. In addition, broad spectrum foragers existed in higher densities and
typically spent a high proportion of time processing food. An increasing use of
plants was believed to induce a local lowering of the overall human trophic level.
This changing strategy led to an increase in dietary breadth, and ushered in the first
experiments in providing food through agriculture and pastoralism.

Zooarchaeologists have since argued that the process of prey switching and
diet broadening occurred far earlier, at the start of the Upper Palaeolithic. Stiner and
colleagues (1999) noted an increased frequency of small fast-moving prey such as
hares, which yield lean meat and are therefore low-ranked, at the start of the Upper
Palaeolithic. This change came at the same time as a reduction of body size in hunted
tortoises, which reflected the increased hunting pressure on this preferred, easy-to-
acquire prey. This contrasted with the pattern of zooarchaeological remains seen
among Middle Palaeolithic Neanderthals in the same geographic regions, which
exhibited a rigid subsistence strategy centred on hunting the most high-ranked of
resources prime-aged medium and large game (Stiner, 1994, 2013; Stiner and Kuhn,
2006).

Stiner (2013) argued the Neanderthal dietary niche changed little in the
hundreds of thousands of years they occupied Eurasia and that it is far easier to find
differences between the Middle Palaeolithic and other periods than within it. This
largely static Middle Palaeolithic foraging niche, narrow and inflexible in most
regions, placed a ceiling on the carrying capacity and ensured a very low population

14
density. In this perspective, as Upper Palaeolithic modern humans entered Eurasia
their broader niche meant they would inevitably displace Neanderthals across all of
their range through a gradual process of competitive exclusion (Stiner et al., 2000;
Hockett and Haws, 2003; Kuhn and Stiner, 2006; O’Connell, 2006; Stiner and Kuhn,
2009).

Comparing plant consumption with the above animal food models is

challenging due to the visibility bias in favour of animal foods. Plant remains are
unlikely to survive on most Palaeolithic sites, thwarting attempts to extrapolate
plant use with conventional approaches, despite being potentially essential for
nutrition (see below) (Speth, 2010). The difficulties faced are exacerbated if data are
viewed in isolation. The dissertation predicts that the contribution of plant foods to
Palaeolithic diets breadth is masked by this taphonomic bias. It hypothesises that it
may be possible to examine the plant contribution to dietary breadth by quantifying
the variety of plant foods represented in dental calculus.

However, before this can take place, a synthesis is needed that re-evaluates
the evidence of plant use in botanical, artefact, genetic and osteological studies to
contextualise this information. This process requires that we first establish what
plant food may have been available to Neanderthals and which were likely to have
been most important from a behavioural ecology perspective.

2.3.3 A behavioural ecology model for Eurasian environments

Although edible energy-rich plants were present throughout western Eurasia

(Sandgathe and Hayden, 2003; Hardy, 2010), there is a dearth of data on their
availability. We can mathematically predict the total biomass of plants in Pleistocene
environments, but not the total biomass of edible plants. Ethnographic data gives us
the option of modelling Neanderthal plant use if they resembled recent foragers and
if Pleistocene environments were similar. Assuming Neanderthals fell within the
ecological gradient present in the economies of recent northern foraging people,
plants would have been a significant part of their diet. Churchill (2014) used net
primary productivity and effective temperature to predict Neanderthal dependence
on plant consumption based on recent foragers. This model estimates that plant
intake represented 11-25 % of diet in the coniferous forests north of the Alps, rising
to 36-43 % in the temperate forest of the last interglacial MIS 5e.

15
Table 1: Energy yields of various food classes consumed by recent foragers. Reprinted from ‘What’s a Mother
to Do? The Division of Labor among Neandertals and Modern Humans in Eurasia’ by Kuhn and Stiner (2006).

N cases kJ/hr kJ/kg

Mean Min. Max. Mean Sd
Large game 4 63,398a 36,000 75,115 6,980b 1,383
Small game 14 16,034a 1,672 56,317 6,980b 1,383
Reptiles 3 15,850 a 17,556 12,435 4,489 b 715
Birds 3 4,472 a 961 8,255
Roots and tubers 14 6,120a 418 26,133 2,926c 1,680
Roots and tubers 9 10412a 3,695 23,333 2,926c 1,680
Roots and tubers 13 1882 d 1,045 2,300 3,136 d 2,338
Seeds and nuts 34 3,520 a 380 18,538 13,188 c 9,334
Seeds and nuts 9 6,508e 1,203 24,933 13,188c 9,334
Seeds and nuts 6 19372d 6,250
Foliage 1,250c 819
Foliage 3 1,534 d 186
Fruits 2,403c 1,463
a Data from Kelly (1995, table 3.3).

b Data from Hawkes et al.(1982), Hurtado and Hill (1987).

c Data from Pennington (1989)
d Data from Wiessner (2003 and personal communication); cases are from Nyae Nyae area minus
those where elephant damage was severe for tubers.
e Data from Wright (1994, table 2).

There is also considerable ambiguity about candidate vegetal food staples and
the types of low- and high-rank foods that we may expect in Pleistocene Eurasia. The
diet-breadth model stipulates that abundance is no predictor of a resources’ value,
meaning that botanical surveys of species frequency are inadequate for extrapolating
staples. There are few studies detailing the returns and costs of acquiring Eurasian
plant foods that would allow us to develop a detailed set of predictions (Martinoli,
2005). Summaries of ethnographic data mostly from Great Basin foragers indicate
that of all classes of plant foods, two categories offer the highest net energy per hour:
seeds/nuts, and underground storage organs (Table 1). Underground storage organ
returns exceed those from seeds/nuts, but they still return less energy than nearly all
classes of animal foods (Kuhn and Stiner, 2006). Researchers have referred to USOs
as a mainly African resource (Kuhn and Stiner, 2006), but this is based on anecdotal
evidence and may not be reliable. Generalisations are inadequate for reconstructing
Neanderthal dietary ecology because understanding plant use requires systematic
data on plant and animal food variables. We know that in some cases prodigious
amounts of USOs were available in Eurasia. For example, reedmace (Typha spp.)
provides extensive and dense concentrations of edible USO biomass in marshes,

16
river and lake shores (Morton, 1975). Unfortunately, detailed ethnographic data of
foragers that specialised in wetlands that we could use to model Pleistocene wetland
foraging is unavailable, because such societies disappeared before they were studied
in detail.

Although there is little information on how the edible plant food biomass
varied in different Pleistocene habitats, we may presume that the availability of
energy-rich plant foods was superior in southern regions. Open Mediterranean
woodland and wetlands would have supported a greater diversity of plant foods
than cold steppe or coniferous forest (Kelly, 1995). In northern areas, although
winters were intensely cold, strong winds often blew away snow, thus exposing
ground-based resources such as edible tubers (Guthrie, 2001). During glacial phases
moisture loving edible plants were drastically limited by the intense aridity but
edible plants tolerant of dry conditions may have been abundant in the rich often-
alkaline steppe soils.

The use of plant foods by recent foragers makes it possible to suggest possible
plant food staples. Diet breadth model criteria (Fig. 2) can indicate whether a food
was high- or low-ranked (Haws, 2004). In Mediterranean woodlands, the high-
ranked plants probably constituted nuts like pistachio (Pistacia sp.), olive (Olea spp.),
chestnut (Castanea sativa), some aquatic plants such as water chestnut (Trapa natans)
and certain underground storage organs including as burdock (Arctium lappa). Mid-
or low-ranked foods likely included seeds of grasses (Poaceae), Madroños (Arbutus
sp.), gum rockrose (Cistus ladanifer) seeds, acorns (Quercus spp.), legumes, and
various woodland, coastal, and wetland underground storage organs (though some
of these foods may have been relatively high-rank) such as honesty (Lunaria annua).
In northern environments, high-ranked foods may have included reedmace,
hazelnut (Corylus avellana), and sea kale (Crambe martima). Lower ranked foods
probably included various underground storage organs like pignut (Conopodium
majus), Asteraceae taproots such as dandelion (Taraxacum), sea holly (Eryngium
maritimum), silverweed (Argentina anserine), and lilies (Lilium spp.), as well as seeds
such as grass grains, and acorns and perhaps pine bark (Pinus spp.). Most of these
foods are relatively energy rich, though many have some processing costs. These
species are far from a complete list of edible Eurasian plants but they are likely
particularly important to Eurasia foragers. Across many of the environments
Neanderthals occupied, a range of other less energy-dense foods were available,
including mushrooms, leafy tissues, stems, drupes, berries, and seaweeds. In
addition to the total energy content, food macro- and micro-nutritional qualities

17
influence forager selection (Hill, 1988). Unfortunately, micronutrient data from wild
plants are extremely rarely compiled and most nutritional data that are available
only covers a fraction of the whole nutritional spectrum So my project focuses on
total energy and macro-nutrients rather than or micro-nutrients.

2.4 The nutritional requirements of Neanderthals

2.4.1 Neanderthal energetics

Distinctive cranial and postcranial morphology distinguishes Neanderthals

from other hominins (Tattersall and Schwartz, 1999). Neanderthals’ anatomy and
inability to reduce the cost of mobility with technology presented energetic
challenges, which may have limited their diet (Verpoorte, 2006; Churchill, 2014).
Overall body proportions show the most resemblance to present-day cold-adapted
populations (Trinkaus, 1981). Scientists interested in Neanderthal energetics have
estimated their body mass range in order to calculate their energetic requirements,
and though these estimates vary depending on the technique and skeletons used, all
versions indicate that the body mass of Neanderthals was very high. Although
Neanderthals stood at a comparable height to recent Arctic forgers, they could have
weighed considerably more (Churchill, 2014). Well-developed muscular attachments
demonstrate that they had heightened muscularity (Churchill and Rhodes, 2006).
Overall, this indicates Neanderthals required increased amounts of energy
compared with other hominins. Estimates of basic daily energy expenditure vary
depending on supposed physical activity level but figures suggest a metabolic
requirement far above the average of any human forager group (Sorensen and
Leonard, 2001; Snodgrass and Leonard, 2009). To fulfil their energy requirements an
ample energetic return from foraging would have been critical (Sorensen and
Leonard, 2001). Recent studies have re-examined Neanderthal energy expenditure
and suggest that although their locomotion required more energy than early modern
humans, the difference is less than previously thought (Hora and Sladek, 2014).
Heyes and MacDonald (2015) have pointed out that the error range in comparisons
between European Neanderthals and Upper Palaeolithic early modern humans is
too great to identify any differences in body mass. However, recent foragers differ
substantially anatomically from early moderns. Our energetic model uses recent
foragers rather than Upper Palaeolithic early moderns as an energetic yardstick.

18
 Animal foods, such as muscle, fat, marrow, and organs, were the most calorie-
rich foods available to Pleistocene foragers. However, there are physiological limits
to animal food diets. If an animal-based diet contains insufficient fat, it can lead to
protein poisoning. Consuming protein beyond a safe threshold leads to a
progressive onset of nausea, wasting and eventual death (Speth, 2010; Butterworth et
al., 2016). Unfortunately, many crucial details of this condition are unknown but
nutritionists have gathered some information from ethnographic sources and small-
scale trials (Stefansson, 1956; Speth, 2010). The notes of explorers and athletic studies
show that people unaccustomed to a high protein intake may rapidly adjust in
weeks to high protein ingestion but a protein ceiling remains (Phinney, 1995; Speth,
2010). Extant hunter-gatherers regularly consume vast amounts of lean protein when
gorging on meat after kills. However, health records suggest that hunters who gorge
in the dry season when game is thin lose weight (Speth, 2010). Recent Arctic foragers
knew the dangers of consuming excessive lean meat and referred to protein
poisoning as rabbit malaise (Stefansson, 1956). Although protein may have formed a
large part of Neanderthals’ dietary intake, a diet of mostly protein is improbable
(Speth, 2010). Nutritionists argue that recent foragers who consume mostly animal
foods overcome these problems by maximising their intake of fat. Neanderthal
hunters must have relied especially on fatty meat to avoid this problem.
Carbohydrates in mammal liver (absent elsewhere due to the effects of rigor mortis),
shellfish and plants could have also played a role (Cordain et al., 2000).

The potential dangers of specialised animal food diets are highlighted in the
case of pregnancy. Hockett (2012) has argued that given the formidable estimated
daily energy expenditure, a diet consisting of only medium- and large-game would
kill a pregnant Neanderthal due to excess protein, vitamin A, niacin, iron, zinc, and
selenium in addition to a major calcium deficiency. This model attempted to take
into account a variety of different levels of activity and thermoregulation, but in all
cases, Hockett argued that a terrestrial animal-food-only diet was toxic to pregnant
females.

2.4.2 Macronutrients provisioning

Along with total protein intake, specific amino acids must be consumed for
maintaining health and ensuring correct development. These nutrients are an
imperative for all mammals and Neanderthals must have acquired them. The body
requires twenty amino acids and about nine must be directly sourced from food. A
19
meat-rich diet would have adequately provisioned Neanderthals with these nine
amino acids (Churchill, 2014). In contrast, sourcing essential fatty acids may have
been more difficult. The essential long-chain polyunsaturated fatty acids
docosahexaenoic acid (DHA) and arachidonic acid (AA) are required for brain
growth. Obtaining sufficient DHA would have been particularly crucial during
pregnancy and lactation (Crawford et al., 2008). The body may synthesise them from
linoleic acid (LA) and α-linolenic acid (LNA), but this is less efficient than acquiring
them directly. Consumers can derive LA and LNA from some domesticated plants
(e.g. flaxseed), but it is unclear if they occur abundantly in wild plant foods
(Simopoulos, 2004). The highest DHA concentrations are found in marine foods.
Foragers could also access this fatty acid through terrestrial mammals, but only in
far lower concentrations, and only in certain tissues like the brain. AA is also present
in terrestrial animals, especially in the viscera, but there are no highly concentrated
sources of AA in large ungulates. Thus, AA deficiencies may have caused problems.
Neanderthals, especially when pregnant, may have required additional AA or DHA.
Supplementary intake of marine fish and mammal brains would be more important
if they consumed a large amount of plant material low in these fatty acids.
Neanderthals may have targeted smaller prey for DHA, as the ratio of brain size to
body mass is larger in smaller mammals (Crawford et al., 2008). This selective
feeding approach would also alleviate the particular risk of protein poisoning in late
winter and early spring when prey had expended their stores of fat over the winter
(Speth, 2010).

We can see that there is a need for a convincing energetic model for
Neanderthals. It is possible that Neanderthals had adaptations that regulated their
essential nutritional demands to what Pleistocene Eurasia could provide, but this is
conjectural. Yet we can deduce without doubt that the large build, muscularity and
large brain that Neanderthals evolved presented energetic and nutritional challenges
although they must have offered some selective advantages. Neanderthal nutritional
demands for fatty acids were the cost of their large and metabolically expensive
brains. This has led Churchill (2014) to argue that the extent to which Neanderthals
relied on plants was constrained by the sheer volume of meat they needed to obtain
essential fatty acids in the absence of fish and shellfish consumption. This hypothesis
is based on incomplete data of the nutritional opportunities of Pleistocene plant
foods. Neanderthals may have alleviated nutritional deficiencies with selective use
of certain resources including plants rich in fatty acids, mammal brains, and marine
and freshwater fish.

20
 2.5 Extant biological and technological traces of diet

Neanderthals were among the first hominins to spread into cold temperate
and glacial environments (Hublin and Roebroeks, 2009), and by studying their
dietary ecology, we may gain a better understanding of the strategies and
adaptations that allowed them to thrive in these inimical environments. We may also
find better explanations of how this close relative survived through 200,000 years of
shifting climate, changing ecologies, and carnivore competition only to abruptly
disappear as early modern humans entered Europe. The gradual expansion of
research using macromammal remains allowed substantial insights into Neanderthal
meat consumption, hunting techniques, and social cooperation (Stiner and Kuhn,
2006; Stiner, 2013; Churchill, 2014). There has been, however, limited ability to
examine their complete dietary ecology. The recent advent and maturation of new
approaches in archaeological sciences such as dental wear, isotope, biomarker, and
dental calculus analyses has allowed considerable strides in building a more
complete view of their dietary ecology. Although these advances have answered
some questions, they have raised others. To explain diet, it is necessary to review
and synthesise this evidence.

2.5.1 The implications Neanderthal technology for subsistence

Hunter-gatherer toolkits can have specialised functions for the collection of

specific resources. Specialised toolkits can reveal resource use if the functional
design can be identified by archaeologists. For instance, later Levantine foragers
produced knapped sickle blades that reflect widespread harvesting of wild grasses
(Bar-Yosef, 1998; Goodale et al., 2010). For these reasons, it is important to review
Neanderthal technology and evaluate if it suggests gathering of plants for food.

Recent hunter-gatherers give us an approximate picture of the tools that

Neanderthals may have used to gather plants. Implements such as stout digging
sticks, folded bark containers, or seed and fruit beaters may have been used, but
these organic items will seldom survive the vast periods of time that has passed. A
handful of wooden implements have survived in special circumstances from 400-125
ka in present-day Germany and England (Thieme, 2000). Mostly these are
interpreted as spears due to their length (1.8- 2.5 m) and pointed tips, however,
digging sticks known from the ethnographic record reach up to 2 m (Nilles, 1942;
Boesch, 2012). Thus, such wooden tools could plausibly have been used as digging

21
sticks, but they would be unusually long examples. Worked sticks resembling
digging sticks in length have also been found at Schöningen (Schoch et al., 2015).
Foraging tools are inclined to be made, used, and disposed over a short period,
leaving few diagnostic markings or use-wear, and thus identification would be
difficult. Furthermore, many of the plant collection tools in the repertoire of recent
foragers are highly multifunctional, for example, foragers can use the same sticks for
prying off edible inner tree bark as they use for dispatching game.

Unlike organic tools, stone tools are regularly preserved in the archaeological
record. Knapped stone tools would have been extremely useful for preparing plant
foods and shaping wood, despite being usually associated with animal butchery
(Langejans, 2012; Hardy et al., 2013; Wadley and Langejans, 2014). Some Middle
Palaeolithic flaked stone tools do preserve evidence of plant processing (Hardy et al.,
2001; Hardy and Moncel, 2011). An alternative proxy line of evidence may be found
in the stone technology used to process food (milling, scraping, and pounding); for
instance, ground stone artefacts may be specialised implements for processing grass
seed (Wright, 1994; de Beaune, 2004; Dubreuil and Nadel, 2015). Like other tools,
they have a generalised function and are used for preparing minerals as well as
processing plants. Ground stone tools for macerating and pulverising have been
widely identified in early Upper Palaeolithic occupations although unequivocal
verification that these ground stones were specifically used for plant processing is
less forthcoming. Ground stones are known in African Middle Stone Age contexts
(McBrearty and Brooks, 2000), but their prevalence is poorly known (Stiner, 2013). In
contrast, ground stones are rarely present in Middle Palaeolithic deposits, but they
are frequently found in Châtelperronian contexts (Straus, 1992; de Beaune, 1993;
Churchill, 2014; Power et al., 2015a). One potential Middle Palaeolithic case was
found associated with pine nuts at Gorham’s Cave in Gibraltar (Barton et al., 1999).
This has been interpreted as a tentative nutcracker. Middle Palaeolithic hominins
may have used modest pieces of naturally shaped stone (manuports) to grind seeds
and nuts, perhaps compatible with their highly mobile lifestyle. Likewise, the cobble
hammers they used for knapping may have been serviceable for plant processing.
Archaeologists might have overlooked such simple toolkits that used unmodified
stones, while the same is not true if they used heavily modified specialised
technology, like in the Epipalaeolithic of the eastern Mediterranean. In sum, we
cannot infer plant use from their technology. Neanderthals could have widely used
grinding stones, but they did not invest in unperishable specialised processing tools.
This appears to be a reliable difference from the technology employed by early
modern humans.

22
 2.5.2 The genetic evidence of diet

The unfolding of genomic data from both living and fossil hominin specimens
has opened a door to a vast array of data on nearly all aspects of hominin biology
(Stoneking and Krause, 2011). With these data, we can test hypotheses about how
hominins biologically adapted to varying diets. The process of piecing together
Neanderthal genetic history has been an incremental undertaking but some insights
into selection relating to diet are already available (Perry et al., 2007). Comparison of
the genomes of contemporary humans, chimpanzees, Neanderthals and Denisovans
has shown evidence that Neanderthals, Denisovans and present day humans all lost
a masticatory myosin gene (MYH16) that helps develop powerful masticatory
capabilities in chimpanzees. Changes in hominin social structure may have
contributed to this but to a large extent this gracilisation is linked to the gradual
adoption of a more energy-dense, softer diet, potentially around ~2 ma (Perry et al.,
2015).

Research has also found that Neanderthals carried activated and deactivated
variants of a gene for a bitter taste receptor- TAS2R38 (Lalueza-Fox et al., 2009).
TAS2R38 detects a compound called PTC (phenylthiocarbamide). PTC does not
occur in plants, but sensitivity to PTC reflects ability to experience bitter tastes in
certain plants such as members of the cabbage (Brassica) genus (Kaplan et al., 1976).
It seems Neanderthals may have experienced variable sensitivity to PTC, as do early
modern humans. Early modern humans, Neanderthals and Denisovans have lost
two genes rating to bitter taste (TAS2R62 and TAS2R64) that are still operative in
chimpanzees. Although in contemporary humans factors such as personal
preference contribute to use of the plants (Niewind et al., 1988) this gene loss means
that they experienced this taste variably as people do today (Perry et al., 2015). Some
report that contemporary humans with the activated TAS2R38 gene eat fewer plants,
but bitter taste receptors may serve to protect against toxins and be quite useful for
identifying plants that are safe to eat. It should be noted that in contemporary
humans non-genetic factors such as personal preference contribute to use of the
plants (Niewind et al., 1988). Low sensitivity to bitterness indicates that
Neanderthals shared specific adaptations with early modern humans associated to
the consumption of particular plant foods and more energy-dense diets. Yet these
pseudogenising (gene loss) mutations occurred long before the divergence between
Neanderthals and early modern humans, and they reflect selection in populations
that predate Neanderthals (Perry et al., 2015). Though we currently have no way of

23
knowing if these adaptations were of dietary importance for Neanderthals, they
probably were not.

Not all selection events that relate to diet predate the Neanderthal divergence
from African populations. Like chimpanzees and humans, Neanderthals possess the
salivary amylase gene (AMY1) enabling them to break down starch into more
useable sugars in the mouth (Perry et al., 2015). Neanderthals, like chimpanzees and
Denisovans, carried only one to two copies of the salivary amylase gene. However,
the contemporary human lineage carries a higher number of copies depending on
the population. Contemporary humans on an average have about six copies of the
AMY1 gene. This difference is thought to have emerged in Africa during the past
200,000 years, long after Neanderthals diverged roughly 600,000 years ago (Perry et
al., 2015). We do not know why copies of AMY1 were selected, since most starch
digestion occurs in the gastrointestinal tract from pancreatic amylase activity (Lee et
al., 2013). Some have proposed that oral starch digestion may have been lifesaving in
infants, which have minimal pancreatic activity (Butterworth et al., 2011; Hardy et
al., 2015a). Extra copies may have arisen to boost protection against death from
diarrheal and intestinal disease in groups heavily reliant on starchy plant foods
(Perry et al., 2007). The high number of copies of AMY1 probably reflects the
importance of starchy plant foods to early African humans. Some might argue that
the fact that Neanderthals had few AMY1 copies implies a low use of plants.
However Neanderthal AMY1 copy number reveals limited insight into their total
intake of plants, because starch is absent in many nutritious plant foods. New World
primates lack AMY1 despite being obligate plant eaters (Perry et al., 2007).

2.5.3 Zoological traces of diet

The vast bulk of data concerning Middle Palaeolithic foraging stems from
skeletal remains recovered from archaeological sites. Faunal remains are far the most
numerous dataset available to researchers, even though many faunal assemblages
are natural accumulations and not a product of hominin activity. This is due to fact
that the karstic caves that dominate Palaeolithic archaeological research in Europe
provide good environments for the preservation of skeletal remains, and act as
landscape bone traps. Anthropogenic macromammal skeletal remains have been
used to target a wealth of questions on meat provisioning capabilities, dietary
breadth, and intensity of resource use. Faunal assemblages are frequently
palimpsests, a sum of many unrelated episodes, such as hunting, scavenging, natural
24
death jumbled by unpredictable sedimentary processes, which leaves a complicated
formation history (Lyman, 2003). Despite the difficulties in reconstructing economic
strategies from skeletal remains, zooarchaeologists have accumulated much
information about Neanderthal- animal interactions. Zooarchaeologists once argued
that Neanderthals were primarily scavengers, due to presumed cognitive or
technical limitations (Binford 1985) but zooarchaeological data have verified that
Neanderthals were capable hunters who exploited a variety of game (e.g. Speth and
Tchernov, 2001). Neanderthal hunting technology is distinct and appears to have
centred on handheld hafted and unhafted spears used mainly for thrusting (Villa
and Soriano, 2010). Specialists argue that Neanderthals’ close range hunting
technology and susceptibility to carnivores meant they may have depended on
closed forests, ecotones, or brush-grass mosaic habitats for much of their kills,
although they clearly ventured into open country to hunt at times (Churchill, 2014).

Neanderthal hunting strategies throughout their range focused on a handful

of key mammals, typically prime-aged artiodactyls. However, Neanderthals were
capable of exploiting most of the herbivorous taxa that they encountered (Churchill,
2014). The largest game available on the landscape is rare in Neanderthal sites, but
there are traces of the consumption of some of these fauna including mammoths (e.g.
Germonpré et al., 2014). They also hunted large, dangerous predators such as bears,
leopards, and cave lions (e.g. Valensi and Psathi, 2004; Blasco et al., 2010). However,
prey exploitation was heterogeneous across their range. Resource choice followed
ecological gradients of the period. On the European Plain fauna that lived in open or
mixed areas such as horse (Equus sp.), woolly rhino (Coelodonta sp.), ibex (Capra sp.),
red deer (Cervus elaphus) and to a lesser extent reindeer (Rangifer sp.) were preferred
(Patou-Mathis, 2000). Yet in Italy, Neanderthals seem to have favoured red, fallow
(Dama dama) and roe deer (Capreolus sp.), suggesting a preference for closed habitats
(Stiner, 1994). In Iberia, a pattern of red deer, horse, ibex, wild boar (Sus sp.) and
bovine exploitation has emerged in the zoological reports although considerable
variability is present. While these data seem to indicate a relatively static hunting
pattern focused on ungulates, there are a few exceptional sites where other distinct
varieties of animals were consumed. Some sites (Figueira Brava, Vanguard,
Gorham's Caves and Grotta di Sant’ Agostino) in Iberia and Italy have evidence for
consumption of monk and ring seals (Monachus monachus and Pusa hispida),
porpoises (Phocoena phocoena) and dolphins (Tursiops truncate and Delphinus delphis)
(Antunes and Santinho-Cunha, 1992; Stiner, 1994; Stringer et al., 2008). Other sites
(Bajondillo Cave, Bolomor Cave and Hayonim Cave) on the Mediterranean rim have
abundant small game components (Cortés-Sánchez et al., 2011; Blasco and

25
Fernández Peris, 2012). Fauna analysts have studied Châtelperronian large-game
fauna assemblages left by late Neanderthals and reported few differences from
Aurignacian assemblages (Grayson and Delpech, 2008), but large game hunting
practises are generally similar between the Middle and the Upper Palaeolithic
(Stiner, 2013).

The role played by small animal prey is far more useful for distinguishing
Neanderthal and early modern human subsistence, but it remains poorly
understood (Fiorenza et al., 2015). The relative lack of information about small game
is due in part to taphonomic problems. Small animals are less likely to be preserved,
recovered, and identified (Yellen, 1991a; b). Foragers may consume small game in
the field rather than bringing them to camp. Small game remains are harder to
conclusively associate with hominin use, because they may be deposited in an
archaeological site by carnivores or birds of prey, or the small animals may have
simply lived, and died, in the site. Furthermore, small game can require less
butchering to process and consume making them even harder to associate with
hominin activity (Brown et al., 2011). Nonetheless, there is sufficient data to
demonstrate that Neanderthals living in southern regions targeted some species of
small game, including rabbits, birds and tortoise (Stiner et al., 2000; Blasco et al.,
2013; Salazar-García et al., 2013). In some cases, such as Grotta dei Moscerini in
central Italy and Hayonim Cave in Israel, small game like shellfish, tortoises, lizards,
and ostrich eggs compromise 45 % or more of the faunal assemblage (Stiner, 1994).
In central and northern Europe, Neanderthals also consumed small game during the
MIS 5e interglacial. In Taubach (Germany) and Vindija (Croatia), there are many cut-
marked beaver (Castor fiber) bones (Gaudzinski-Windheuser and Roebroeks, 2011).

Fish bones have been found in several sites in western Italy, France, and
southern Iberia (Fiore et al., 2004; Fiorenza et al., 2015), as well as Vindija Cave in
Croatia (Paunović and Smith, 2002) and at Raj Cave in Poland. Like small terrestrial
game and shellfish, fish remains are rare in Mousterian levels of Palaeolithic sites.
Even where fish remains are present, they are less frequent than in the
contemporaneous Middle Stone Age sites in Africa (Klein and Steele, 2008), but this
disparity may be related to sea level changes, as many western Middle Palaeolithic
coastlines are currently under water and thus not available for study. In addition,
fish remains are even less likely to be preserved than terrestrial small game (Szpak,
2011). A variety of shellfish species were found in Middle Palaeolithic sites in Iberia,
Italy and Greece, but they are very rare (Stiner, 1994; Klein and Steele, 2008).

26
 The use of small game and aquatic resources directly relates to dietary
breadth. Zooarchaeologists usually consider small game a low-ranked prey item,
because the costs are high relative to the amount of food each prey item provides.
For example, a hunter may expend considerable energy to bag a fast-moving and
agile hare with limited energetic returns, unless they use technology to assist the
process. Dependence on low-ranked prey is often linked to the declining supply of
high-ranked large and medium-sized game, population growth, and technological
investments in energy capture (Stiner et al., 2000). However, regarding all small
game as low-ranked prey is overly simplistic (Fiorenza et al., 2015). Some small prey,
such as tortoises and eggs, may yield high returns with little foraging costs. In other
cases, small prey may be low-return but may offer specific macronutrients (e.g. DHA
in marine foods or small mammal brain) more important than total energy (Kelly,
1995; Winterhalder and Smith, 2000; Haws and Hockett, 2004). For these reasons,
zooarchaeologists have attempted higher resolution approaches including sub
classifying small game by ease of capture and species diversity (Stiner, 2001). These
dietary breadth metrics have been used to argue that Neanderthals very rarely
captured low-ranked small game, in contrast to recent hunter-gatherers, some Upper
Palaeolithic foragers, and possibly some Middle Stone Age foragers (Stiner, 2013).
However, there is still some apparent variability in Neanderthal behaviour. At
Kebara Cave, Neanderthals increased their reliance on low-rank juvenile gazelle and
fallow deer (low-rank due to their small body size and reduced adipose tissue),
while the relative proportion of high-ranked aurochs, red deer, and boar decreased
from 50 ka onwards. In this case, it appears that Neanderthals depleted large game
supplies, and were forced to adapt through prey switching (Speth and Clark, 2006).

In sum, Neanderthals appear to have been capable hunters who favoured

hunting medium and large game. The absence of small game, fish, and shellfish in
their range in central and northern western Eurasia is due to a combination of bias in
the archaeological signal and intentional hunting strategy. Although the Middle
Palaeolithic predation niche varied considerably through time and space, the general
pattern of Neanderthals as medium- and large-game hunters is to some extent
correct. However, in some cases they were also consumers of other foods such as
small mammals, fish, shellfish, bird eggs, lizards, and scavenged meat.
Unfortunately, researchers have yet to describe full geographic and temporal
variation in small-game procurement (Fiorenza et al., 2015). Nevertheless, the
majority of studied faunal assemblages indicate most Neanderthals appear to have
engaged in narrow spectrum foraging for most Neanderthal populations, with some

27
evidence for increased dietary breadth beginning about 50 ka (Speth and Clark, 2006;
Stiner, 2013).

2.5.4 Indications of diet from macrobotanical plant remains

Evidence for the consumption of plant foods is sparse across the Neanderthal
range in part because, unlike bones, most plant remains require a specific set of
exceptional circumstances to preserve in archaeological deposits. In the western
Eurasian context, this is typically carbonisation, though desiccation and
waterlogging are also possible (Van der Veen, 2007). Carbonisation requires that the
food plants are exposed to fire and typically best preserves seeds and nuts that
benefit from cooking. The record of charred remains is biased in other ways too,
because foragers frequently consume plant foods as they are collected, before
returning to camp and encountering fire (Marlowe, 2010). In addition,
macrobotanical assemblages poorly preserve tissue of some vegetal resources such
as underground storage organs, leafy plant parts and oily plant foods (Pennington
and Weber, 2004), rendering these foods largely invisible to archaeologists. Plant
remains cannot be representatively recovered or even detected onsite without
specialised archaeobotanical sampling, which the archaeobotanical community only
commenced in the 1960s. Therefore, and unlike studies of fauna, literature so
scarcely discussed botanical remains that some archaeologists may have avoided
monitoring for them even after the techniques were available. In addition, even if
plant remains are found, they may simply reflect plants growing near the site rather
being a signal of food items. Neanderthal sites with plant remains are rare but this is
a preservation bias.

Table 2: Neanderthal sites with evidence of macrobotanical plant remains.

Site Macroremains Region Complex Reference

Douara Cave Hackberry Syria Mousterian Matsutani, 1987; Griggo, 2004
Gorham’s Cave olive, stone pine Iberia Mousterian Barton et al., 1999
Rabutz hazelnut Germany Mousterian Toepfer, 1958
Mas-des-Caves hackberry France Mousterian Barton et al., 1999
Kebara Cave pistachio, grasses, lentil and other Israel Mousterian Lev et al., 2005
legumes
Theopetra Cave lentil, chickpea and other Greece Mousterian Mangafa, 1998
legumes, grasses and nuts

28
 The few existing Middle Palaeolithic sites with botanical remains provide a
varied but are incomplete for all inhabited environments (Table 2). At Douara Cave
in Syria, abundant deposits of hackberry (Celtis sp.) were identified dating to
roughly 40 to 55 ka (Matsutani, 1987; Griggo, 2004) Barton and colleagues also
reported this plant taxon from Mas-des-Caves in France (1999). Archaeobotanists
have found macroremains from stone pine nut (Pinus pinea) and olive (Oliva spp.)
dating to 51,700+3300 BP 14
C in the western Mediterranean at Gorham’s Cave,
Gibraltar. Charred hazelnut shells (Corylus avellana) have been identified at Rabutz
in central Germany during the warm conditions of the MIS 5e interglacial (Toepfer,
1958), although these shells could have entered the archaeological site by natural
processes. Lentil, chickpea, pea and vetchling are reported from Middle Palaeolithic
deposits at Theopetra Cave (Mangafa, 1998). The most notable and diverse
macrobotanical assemblage so far identified was located in Kebara Cave at Mt
Carmel (Israel). This assemblage of charred seeds dates to 63-48 ka, and dwarfs in
diversity and size all other Middle Palaeolithic as well as many Upper Palaeolithic
macrobotanical assemblages so far recovered. The contents suggest a broad foraging
strategy for potential staples, mostly legumes (Fabaceae) with some acorns (Quercus
spp.), pistachio nuts (Pistacia spp.), and chenopods (Lev et al., 2005). Pistachio nuts
are rich in lipids, proteins, and carbohydrates, and are therefore an excellent
candidate high-rank food, although they would have been available only for a brief
season (Dreher, 2012). However, the richness of the legume assemblage is unusual,
although protein-rich, they are slow to collect and are arguably low-ranked foods
and more usually associated with near-sedentary Epipalaeolithic groups of the Near
East (Savard et al., 2006). Many plants were highly restricted by season and would
have to be harvested from different habitats in windows from spring to autumn.
Overall, the plant remains present evidence of plant use across a variety of
Neanderthal habitats, and in at least one case, there are traces of particularly broad
use of plants. Yet most macrobotanical examples cannot be unambiguously
associated with diet. The collated macrobotanical evidence is promising but overall
data are too fragmentary to explore the variation of plant use.

2.5.5 Sedimentological traces of microbotanical plant remains and diet

It is possible to collect data about subsistence patterns from particles in

archaeological sediments. Phytoliths, also known as plant opal, are glassy bodies
comprised of biogenic silica found in the aerial tissue of plants (Piperno, 2006). They

29
often preserve specific morphologies relating to the plant taxon or plant part that
produced them. The decay of the plants releases the phytoliths and thus they enter
the archaeological record. Phytoliths routinely survive for vast spans of time and can
survive for hundreds of millions of years in certain conditions (Carter, 1999). Thus,
they readily survive in the sheltered conditions of cave sediments (Albert et al.,
1999). However, if stratigraphic levels are mixed and poorly associated with
archaeological evidence it may be difficult link causality of phytoliths to hominin
activity. Phytoliths can enter archaeological sites from windblown aerosols,
colluvium, bird droppings and other animal activity. Phytolith studies may require
well-preserved archaeological deposits examined collaboratively with FTIR and
micromorphology. If properly performed, phytolith analyses can provide detailed
information about the use of specific plants for many different uses. So far, it has
been possible to infer food, bedding, or fuel using phytolith assemblages. Phytolith
specialists have studied only a handful of Neanderthal sites for phytoliths. At 58-37
ka cal BP levels in Esquilleu Cave in Cantabria (Iberia), phytoliths indicate
continuous deposition of grass leaves by a hearth, suggesting the presence of a
bedding zone at this spot (Cabanes et al., 2010). Phytoliths at Amud cave in Israel
also indicate plant bedding dating to 70-55 ka (Madella et al., 2002). In some cases,
such as Kebara Cave, analysts have retrieved phytoliths from hearth deposits and
inferred fuel choice (Albert et al., 2000). Even more interestingly, this site also
contained high concentrations of the dendritic morphotype phytolith, which would
usually be more familiar in agricultural contexts due to the abundance of
domesticated grain. The authors interpreted this accumulation as evidence that
Neanderthals repeatedly collected mature grass seed. This is a controversial
interpretation because the same pattern could be the result of fauna burying caches
of seed. If the anthropogenic interpretation is accepted, it may suggest that
Neanderthals in one Levantine site made heavy use of a low-ranked food.

New studies have explored the potential for detecting other anthropogenic
indicators in archaeological sediments. Researchers have highlighted that the
products of biological processes (biomarkers) may yield insights into the dietary
inputs of hominin metabolisms. Most analyses so far have focused on sterols and
stanols as candidate faecal biomarkers, because they have the virtue of high stability
through the food chains and are resistant to diagenesis (Peters et al., 2005). Only
higher mammals form 5β-stanols, which they produce in their intestinal tracts
during the metabolic breakdown of cholesterol and phytosterols. Their use as an
anthropogenic indicator relies on identifying the source coprolite as hominin.
Fortunately for archaeological science, the relative proportions of these stanols and

30
sterols are known to be indicative of dietary preferences although how this works is
not understood (Floate, 1970; MacDonald et al., 1983). Sistiaga and colleagues (2014)
took sediment samples from morphologically identified coprolites near combustion
features in the open-air site of Abric d'El Salt (Alicante, Iberia) dated to 60.7 ± 8.9 and
45.2 ± 3.4 ka (Garralda et al., 2014). These samples were analysed with gas
chromatography- multiple reaction monitoring-mass spectrometry. Some samples
were dominated by coprostanol and its diagenetic product epicoprostanol, verifying
that the samples represented coprolites. The ratio of coprostanol and phytosterol can
indicate which taxa produced a coprolite. In all cases at El Salt the values were high,
so the authors argued that the faecal residues are from suids or humans, and because
no suids were found at the site, humans were the ostensible producers (Sistiaga et
al., 2014). The authors report that they discovered from this find that Neanderthals
have a high rate of conversion of cholesterol to coprostanol, yet in a fallacy of
circularity, use this same trait to identify the coprolite as Neanderthal (Sistiaga et al.,
2014). The faecal biomarkers almost certainly represent many separate events and
thus cannot be identified to hominin without further biomarkers (i.e. bile acids) (Bull
et al., 2002). Yet, although results from El Salt do not reveal diet, multi-pronged
faecal biomarkers approaches will likely be highly useful, especially if coprolites
deposits are discrete and can be unambiguously identified as Neanderthal.

2.5.6 Evidence of palaeodiet from pathologies

Health is deeply interrelated with diet and many disorders are a result of food
choice. In some cases, details on health can be gleaned from surviving hominin
skeletal material. Carious lesions are cavities that form on the surface of tooth
enamel from the demineralising effects of oral microbiota (Selwitz et al., 2007). The
intake of carbohydrates is one of several factors that are needed to form carious
lesions (Selwitz et al., 2007) and thus carious lesions can be indirect evidence of diet
in contemporary humans. The frequency of these lesions increases relative to the
amount of carbohydrates consumed, leading many archaeologists to use this as a
proxy for palaeodiet (e.g. Christophersen and Pedersen, 1939; Larsen et al., 1991;
Flensborg, 2011). Compared with typical contemporary people, caries are rare in
recent foragers (<10 %), but common in past agriculturalist groups (Lanfranco and
Eggers, 2012). Dental caries are very rare in surviving Neanderthal teeth (Tillier et
al., 1995; Walker et al., 2011b). Of the approximately 1250 Neanderthal teeth
examined, just six (0.48 %) have been reported to display carious lesions. Of the six

31
cases, three occur in Iberia, two in France and one in Israel (Lanfranco and Eggers,
2012). Yet seldom do researchers identify caries in early modern humans, although
their frequency is unclear. Nevertheless, the rarity of Neanderthals suffering from
caries appears to reflect a reliance on animal foods. Not all agree, and the dearth of
caries in Neanderthal teeth is taken by some to reflect absence of cariogenic bacterial
species in Neanderthal oral flora (Sołtysiak, 2012). For example, molecular evidence
suggests that cariogenic bacterial species Streptococcus mutans was not present in
archaic humans. Yet microscopic examination of dental calculus identified these
bacteria from the Kebara 2 and the Subalyuk 1 Neanderthals. In addition, caries may
arise from the activities of other cariogenic bacteria species (Tomczyk, 2012).

Another example of a pathology demonstrating diet is dental calculus. The

formation of dental calculus is a multicausal process (See 2.6.1), but some dietary
factors are clearly involved. Protein consumption enhances calculus formation by
increasing the pH of the mouth (Lieverse, 1999). Little effort has been made to
extrapolate diet from the abundance of Neanderthal calculus deposits. This is
unsurprising as interpreting diet from calculus abundance is challenging. Some
agricultural populations with low protein intake have high calculus abundance
while some forager populations with ample meat use have little calculus (Lieverse,
1999). This pattern has led some palaeopathologists to use high calculus abundance
to trace a high use of carbohydrates instead of protein (Greene et al., 2005). Saliva
flow, silicon intake, smoking and predisposition acerbate dental calculus formation
in addition to diet explaining this contradiction (Bergström, 1999; Lieverse, 1999).

2.5.7 Evidence of diet from dental wear

Over the course of life, wear reduces the surfaces of teeth. Wear is heavily
influenced by the mechanical properties of the food consumed and thus may reveal
information on the characteristics of diet (Ungar, 1998). Food may have varied
physical properties, such as abrasiveness, toughness, hardness, and brittleness,
meaning different foods requires different masticatory processing (Cromton and
Hiiemae, 1970; Fiorenza et al., 2011). Over time, attrition, abrasion, and erosion
combine to gradually remove the enamel surface of teeth (Kaifu et al., 2003; Addy
and Shellis, 2006; Kaidonis, 2008). Attrition is the mechanical force exerted from
contact of opposing teeth. Abrasion is another physical wear caused by the rubbing
of exogenous material pushed against teeth during mastication. Particles in food
such as phytoliths that are softer than enamel are thought to still wear enamel
32
because they force apart the proteins that hold enamel crystallites together (Xia et al.,
2015). In the context of Pleistocene foragers, this exogenous matter is predominantly
hard and fibrous foods, foreign particles transported in food, and environmental
dust carried in wind. Erosion is the chemical dissolution of the tooth surface, but its
incidence in the teeth of foragers is insignificant, whereas abrasion and attrition are
commonplace (Fiorenza, 2015). Thanks to the gradual advances in our knowledge of
the mechanisms of how dental surfaces reform over life, dental wear has matured
into a widely applicable means of dietary reconstruction. As the discipline has
grown, dental wear has been used to broadly classify the diet of recent forager
groups (e.g. El Zaatari, 2010) and the feeding niches of ancient hominins (e.g. Ungar
and Sponheimer, 2011). The discipline comprises of two main fields based on the
nature of the wear studied - microwear and macrowear.

Dental microwear analysis is the study of the microscopic damage on a tooth's

surface as the result of its use. The surfaces of many Neanderthal teeth are highly
worn, and this necessitated that early microwear research used the striation patterns
on cheek side the buccal surface as a proxy for food masticated on the chewing
surface (occlusal surface) (Pérez-Pérez, 1994; Lalueza et al., 1996; Puech, 1999). The
buccal wear pattern in individuals consuming a high proportion of meat is
characterised by a lower number of striations and a relatively high proportion of
vertical ones, while individuals relying on a more vegetarian diet display an
increased number of striations, with a greater proportion of horizontal ones (Fox and
Pérez-Pérez 1993; Pérez-Pérez 1994). Using scanning electron microscopy, wear
specialists (Lalueza et al., 1996) compared Neanderthals with archaic Homo
specimens and with samples of recent people. Their samples of recent people
represented strict vegetarians, tropical and subtropical foragers who consumed
relatively high proportions of plant foods, and high-latitude foragers and
horticulturalists who consume large amounts of meat. The eight sampled
Neanderthals (La Quina V, Gibraltar 2, Tabun 1 and 2, Amud 1, Malarnaud, Saint
Césaire and Les Pradelles) had different microwear patterns from those of
vegetarians. Some Neanderthals were similar to living groups with high-meat diets,
but most fell within the range of both high-meat and more mixed diets. Furthermore,
the results taken as a whole showed no compelling chronological, climatic, or
geographic patterns.

Buccal and occlusal wear both reflect diet but emerge in different patterns.
The formation of occlusal wear is better-understood than buccal wear, making it a
more informative approach for ancient hominins (El Zaatari, 2007). More recently,

33
wear studies have grown in sophistication with the integration of confocal
microscopy with the advent of "occlusal microwear texture analysis". El Zaatari
(2010) used this approach to analyse the occlusal microwear of recent foragers from
known temperate, arctic, and other biomes. The groups had varying amounts of
marine foods, large game, small game, and plant foods. These samples provided the
baseline to which to compare 35 Neanderthal individuals from 23 European and
Levantine sites. By grouping these sites by habitat type, El Zaatari showed that
Neanderthals associated with mixed and wooded environments consumed plant
foods, but in a lesser quantity than meat. Four Neanderthals deriving from open
habitats (Spy, La Quina, Arcy-sur-Cure, and Subalyuk) most closely resembled
recent groups who fed on fish, seals and guanaco (Lama guanicoe) with few plants
(about <15 %) (El Zaatari, 2010; El Zaatari et al., 2011). The eight Neanderthals from
mixed habitats (Saint Césaire, Petit-Puymoyen, Rochelot, La Chaise, Vindija, Kebara,
and Tabun) more closely clustered with marine, small game and plant consuming
foragers. The four Neanderthals from closed habitats (Amud, El Sidrón, Grotta
Breuil and Zafarraya) had higher levels of molar occlusal complexity and
heterogeneity indicating the highest levels of plant use of all the Neanderthal
samples. Despite this, they did not cluster with the forest-dwelling recent foragers,
tantalisingly suggesting they may not have been as reliant on plants.

The other main approach to wear-based dietary reconstruction is dental

macrowear. Molar macrowear represents the cumulative impact of the mechanical
properties of diet during an entire lifetime, unlike dental microwear which covers
only a brief period just prior to death (Grine, 1986; Janis, 1990). Early efforts to
document dental wear interpreted the generality of the abrasiveness of diet rather
than the components of diet (Fiorenza et al., 2015). The effectiveness of macrowear
studies for dietary interpretations has greatly grown in recent decades with the
improved knowledge of how mastication reforms occlusal contact areas (Douglass
and DeVreugd, 1997). These developments led Kullmer and colleagues (2009) to
develop 3D virtual models to analyse wear patterns on the facets of teeth. By
measuring perimeter, inclination, and orientation, it became possible to create a
model of how food was chewed, in a method termed "occlusal fingerprint analysis"
(Kay and Hiiemae, 1974; Janis, 1990). Fiorenza and colleagues (2011) analysed the
occlusal fingerprints of 19 Neanderthals, which he grouped into a deciduous
woodland group (Krapina), a steppe and coniferous forest group (Monsempron, Le
Moustier and Vindija) and a Mediterranean evergreen woodland group (Amud,
Tabun and Shanidar). Wear patterns of the deciduous woodland and Mediterranean
evergreen woodland was suggestive of a mixed diet probably containing a

34
significant amount of plants. This group matched wear observed in several plant-
reliant recent human reference populations. Neanderthals from steppe and
coniferous forest regions exhibited patterns of cold dwelling groups that consume
tough foods such as terrestrial mammal muscle or marine foods, though it was not
possible to distinguish between these two (Fiorenza et al., 2011). More recently,
occlusal fingerprint analysis has explored wear patterns of Neanderthal molars
found on the Italian peninsula, Saccopastore 1 and 2, and Guattari 2 and 3 (Fiorenza,
2015). This study found wear suggestive of the use of animal and plant foods on all
specimens, though the Saccopastore specimens were more suggestive of meat eating.
Guattari 2 fell closer to the previous Mediterranean reference group and Guattari 3
clustered together with the deciduous woodland reference group. Fiorenza and
colleagues (2015) interpreted these results as suggesting plant foods had a degree of
importance in the warm interglacial MIS 5, while during warm phase MIS 3
Neanderthals appear to have relied more on animal foods.

Micro and macrowear provide a substantial amount of information of the

mechanical characteristics of masticated food. From these data, inferences on the
nature of diet can be made. However, interpreting these data presents significant
problems relating to comparability of environments and differences in reference
populations. For instance, the lumping of steppe with coniferous forest biomes
makes comparison difficult (Fiorenza et al., 2011). Furthermore, some dental wear is
derived not from diet but foreign mineral particles that come in to contact with the
enamel, which can confound wear studies (Lucas et al., 2013). Nonetheless, wear-
based approaches provide information about Middle Palaeolithic diets unavailable
from other lines of evidence. These lines of evidence suggest that Neanderthals
consumed plant foods as part of a diet rich in animal foods, and that there may have
been more variability across habitats than isotope studies indicate.

2.5.8 Isotopic approaches to palaeodiet

The maturation of methods of dietary reconstruction borrowed from other

fields has made major contributions to understanding Palaeolithic diets. The
application of stable isotopic analysis to ancient hominins is one example that has
become an important means for reconstructing diets and corroborating other lines of
evidence (e.g. Codron et al., 2008). Stable isotopic palaeodietary analyses principally
use carbon and nitrogen isotopes from collagen from bone and tooth dentine.
Isotopes from other elements such as sulphur also can reveal dietary history (Privat
35
et al., 2007). Stable isotopes are analysed as the relative amount of a heavier isotope
to a lighter isotope and expressed in δ notation in parts per mil (Schwarcz and
Schoeninger, 1991). Carbon isotopes (13C/12C) can provide information about the
consumption of plants and marine foods, and the nitrogen isotopes (15N/14N) reflect
use of plants and animals and trophic level. Stable isotopes can sometimes provide
detailed information on how much consumed protein was from terrestrial animals,
marine animals, freshwater animals or plant foods. Comparison of values with
contemporary fauna serves as a reliable means of quality control. The isotopic signal
of hominin bone collagen reflects a variable amount of time but given its turnover
rate it reflects years of diets in adults (Hedges et al., 2007).

Studies of Neanderthal diets using stable isotopes have garnered much

attention. This analysis requires well-preserved collagen and thus is limited by
chronology and the taphonomic conditions present at a site. Values have been
published from at least 22 individual Neanderthals from 14 sites (See Table 3), and
ambiguous values from two others. These sites date between 120 and 30 ka, and are
located in France, the Netherlands, Belgium, Russia, Germany, and Croatia
(Richards et al., 2000; Bocherens et al., 2005; Beauval et al., 2006; Krause et al., 2007;
Richards and Schmitz, 2008; Hublin et al., 2009; Ecker et al., 2013; Wißing et al.,
2015). Though isotope and fauna bone studies conflict on Neanderthal prey choice,
both indicate that the protein in Neanderthal diets came predominately from
terrestrial animals, likely medium and large herbivores (Dusseldorp, 2010, 2013).
This isotopic signature has been interpreted as largely representing a consistent
ecological niche at the apex of the Pleistocene terrestrial food chain. Isotope values
from fauna bones from the same sites suggest that Neanderthals were higher on the
food chain than even carnivores such as wolves and bears, but in some cases the
comparative fauna was from different archaeological levels and thus may not
provide a reliable baseline (Richards et al., 2000, 2008; Bocherens et al., 2005). Some
researchers have gone as far to say that isotopically, Neanderthals mimic obligate
carnivores (Churchill, 2014). Notably the Saint-Césaire Neanderthal associated with
a Châtelperronian tool kit does not differ from the other Neanderthals despite the
other indications from this technology that suggest more reliance on plant foods (See
2.5.1). Furthermore, the isotopic values give no indication that Neanderthals
consumed aquatic resources like fish or shellfish. The sampled individuals were
mostly from inland regions where marine foods are not expected, but the absence of
freshwater fish is surprising. Similarly, the Neanderthal isotopic signature appears
to leave little room for consumption of plant proteins. However, it cannot entirely
rule out a regular intake of protein from plants, due to differences in absorbable

36
protein compared with meat. Plant nutrients such as protein and lipids are often less
absorbed than animal equivalents (Baer et al., 2012). Plant foods are typically high in
carbohydrates and often contain only moderate levels of protein.

Table 3: Neanderthal remains with published stable isotopic values (δ 13C and δ15N).

Site N Age Region Predominant diet Source

Payre a 1 MIS 8-7 France Terrestrial animal Ecker et al., 2013
Scladina 2 MIS 5c-b Belgium Terrestrial animal Bocherens et al., 1999, 2005
Les Pradelles b 5 MIS 4 France Terrestrial animal Beauval et al., 2006; Bocherens et
al., 2005
Okladnikov 1 MIS 3 Russia Terrestrial animal Krause et al., 2007b
Vindija 2 MIS 3 Croatia Terrestrial animal Richards et al., 2000
Feldhofer 2 MIS 3 Germany Terrestrial animal Richards and Schmitz, 2008
Les Rochers- 1 MIS 3 France Terrestrial animal Beauval et al., 2006
de-Villeneuve
Saint-Césaire 1 MIS 3 France Terrestrial animal Bocherens et al., 2005
Jonaz 1 MIS 3 France Terrestrial animal Richards et al., 2008c
Goyet 12 c MIS 3 Belgium Terrestrial animal Wißing et al., 2015
Spy 1 MIS 3 Belgium Terrestrial animal Bocherens et al., 2001
Engis 1 MIS 3 Belgium Terrestrial animal Bocherens et al., 2001
Zeeland Ridge 1 MIS 3 Netherlands Terrestrial animal Hublin et al., 2009
a This sample derives from tooth enamel and thus has only carbon isotopic values.

b The reliability of the isotopic values has been questioned by Bocherens et al., 2005

c The twelve samples represents four or more individuals.

Neanderthals may have plausibly targeted vegetal foods low in protein and
high in carbohydrates and lipids to ameliorate the risk of protein poisoning (See
2.4.1). We cannot quantitatively estimate contributions of each component of diet
unless a mathematical model (mixing model) is used (Bocherens, 2009; Fernandes et
al., 2014). Reliably fitting of such models requires the input of isotopic values of all
the consumed foods, and this is not possible in a Palaeolithic context. Therefore,
mixing models in these contexts might be misleading.

Some have challenged the view that Neanderthal protein intake was near
exclusively animal-based. Specialists have explored various possibilities to assess if
the nitrogen isotopic values reflect an extremely high trophic level. Speth ( 2010)
noted that severe nutritional stress can lead to increases in δ15N due to the effects of
protein catabolism, and that this starvation signature may explain the elevated
Neanderthal δ15N signal. Bouts of starvation are a well-documented part of life for
some foragers, particularly those in high latitudes such as Arctic foragers. Episodes
of stress (nutritional or illness related) endured by Arctic foragers in childhood are
visible with enamel defects such enamel hypoplasia on their teeth. Comparison of
Arctic foragers and Neanderthal indicate comparable stress levels (Guatelli-
Steinberg et al., 2004). However, given the slow turnover rates of bone, malnutrition

37
severe enough to elevate δ15N is would normally be fatal before it could be recorded
in bone (Beaumont et al., 2013), suggesting that malnutrition might not explain the
Neanderthal signal.

Another source of confusion may come from the fauna used as a baseline for
extrapolating Neanderthal trophic level. If prehistoric faunal diets differed from that
of their present day counterparts, it may confuse our interpretation of Neanderthal
diets. For example, if the herbivores that Neanderthals consumed had elevated δ 15N,
then the Neanderthal trophic level would appear high. This would mask
consumption of plant protein if comparative fauna were unavailable. Usually ideal
numbers of comparative fauna from the same levels are absent from isotopic studies.
Furthermore Eurasian elephantids have unusually high δ15N values unlike present
day elephants (likely relating to consumption of faeces), and some have suggested
that this explains the apparent trophic level of Neanderthals (Richards et al., 2000;
Bocherens et al., 2005; Kuitems et al., 2012). However, elephantid consumption is
unlikely to explain high δ15N values for all individuals published. Neanderthals
consumption of elephantids appears to have been rather limited (See 2.5.3), while the
Neanderthal isotope values are remarkably consistent across samples. The
consumption of high δ15N nitrogen prey may have had an impact, but the potential
to distort Neanderthal isotopic signals should not be overstated in this case.
Although isotopic studies have given a powerful insight into protein consumption,
these sampled Neanderthals are disproportionately from northern open
environments and cold phases (Richards and Trinkaus, 2009; Salazar-García, 2012).
Few southern Neanderthal isotopic values have been published, in part due to the
poorer preservation of Neanderthal collagen in these warmer climates (e.g.
Ambrose, 1990). Of the published 22 Neanderthals subject to collagen isotopic
studies, only two individuals lived in the forested interglacial period (MIS 5), while
others are derived from a range of environments from climatic phases. This biased
sample should temper interpretations based on isotopic data and reveal little about
dietary variation in different environments.

2.5.9 The contribution of dental calculus to understanding Middle Palaeolithic diets

One the most exciting emerging ways to learn about ancient diets is to use
hominin dental calculus. Dental calculus along with dental enamel is the only tissue
in the human body with no means of regulated shedding. This unique characteristic
enables entrapment and preservation of food particles and other materials become
38
entrapped in this biomineral deposit following consumption. Such an approach
could address how Neanderthals used plants. Dental calculus (tartar) is dental
plaque that has become calcified by salivary calcium phosphate (Lieverse, 1999). It is
a ubiquitous pathology of the mouth in humans and human relatives. Researchers
have reported finding starch grains, phytoliths, pollen, diatoms and other particles
relevant to life history entrapped in human dental calculus for extended periods of
time (e.g. Dobney and Brothwell, 1988; Boyadjian, 2012). Dental calculus sampled
from living or dead individuals is rapidly gaining recognition as an invaluable
material for the reconstruction of life histories. The integration of dental calculus
analyses to Palaeolithic hominin remains has made powerful contributions to our
knowledge of Neanderthal diets. As dental calculus offers direct evidence of the
plants that entered the mouth, this analysis can potentially give information of plant
use that is invisible with other methods.

Henry and colleagues (2011) pioneered the application of this approach to the
elucidation of Neanderthal diets. This study used dental calculus sampled from one
individual from Shanidar Cave, Iraq, and from two Neanderthal individuals from
Spy Cave, Belgium. These analyses recovered remains of phytoliths from date palms
(Phoenix spp.) and starches from grass seeds (Triticeae), legumes (Faboideae) and
potential indeterminate underground storage organs. Although the assemblages
probably reflect consumed foods it is difficult to rule out the contribution of chyme
(semi-digested stomach contents). Chyme is widely consumed by foraging societies
and it probably was a feature of Middle Palaeolithic diets (Buck and Stringer, 2014).
However, starches predominated in these samples, yet ungulate chyme would
predominately contain phytolith not starches, arguing against chyme being the
primary source of plant remains in dental calculus.

Some of the starch grains, including Triticeae starches, were apparently

partially disrupted (semi gelatinised). Gelatinisation is a process where starch
undergoes a breakdown of its intermolecular bonds when heated in the presence of
water. Thus, semi gelatinisation was interpreted as evidence of the controlled boiling
and cooking of these starches. Yet some have queried whether this process could
occur spontaneously to starches trapped in calculus for tens of thousands of years
(Collins and Copeland, 2011). Current research is investigating the tempo of
spontaneous gelatinisation and it remains to be seen if we can ascertain if
Neanderthals boiled Triticeae seeds.

Another study sampled dental calculus Neanderthals from El Sidrón in

northern Iberia, where a rich assemblage of Neanderthal remains was found. Hardy
39
and colleagues (2012) analysed dental calculus from five Neanderthals dating to
51,100 ka. Notably, this sample is probably the only example from temperate
woodland environment so far published. Hardy recovered moderate numbers of
starches in four of the five samples and one grass phytolith in one of them using
optical microscopy. In addition, the dental calculus samples were analysed with
(thermal desorption and pyrolysis) gas chromatography-mass spectrometry to assess
if compounds relating to diet could be present. This technique yielded molecular
evidence of inhalation of wood smoke and naturally occurring bitumen. The
presence of compounds such as chamazulene, dihydroazulene, and 4-
methylherniarin suggested exposure to yarrow and chamomile herbs. Hardy
interpreted this as traces of plants used for medicinal rather than for nutrition
purposes. Critics have raised concerns that these herbs may enter diet in chyme
(Buck and Stringer, 2014). Pleistocene ungulates probably commonly grazed on
yarrow. In addition to undeliberate use in chyme, plants like yarrow and chamomile
may have been gathered and consumed as a vegetable. Although bitter-tasting, Buck
and Stringer (2014) point out that traditional Alaskan people consumed camomile as
a food plant (Kuhnlein and Turner, 1991). All food types may enter the mouth in
many alternative ways such as ritual uses, dental hygiene or accidental intake.

Later published research has attempted to address the lack of data on

Neanderthals from Mediterranean environments. Within a multi-proxy research
article about Neanderthal diets from eastern Iberia, the author analysed dental
calculus from nine teeth and tools from Sima de las Palomas del Cabezo Gordo in
southeast Iberia (Salazar-García et al., 2013). To control for contamination, wash
samples of cave rock fall (éboulis) were collected, and these showed grass phytolith
and some starch contamination traces. However, these control stones were selected
from a unit balk exposed to atmospheric airborne microremains for an extended
span of time. Calculus and sediment adhering to fauna teeth from the site were also
examined, analysis found no starches and only three phytoliths, mostly coming from
grasses. Unlike the controls Neanderthal calculus recovered microremains types
included leafy matter indicated by polyhedral multi-cells, hard endosperm of seeds
or nuts as well as grass seeds and possibly underground storage organs. The starch
grains found in dental calculus samples largely overlapped with the types recovered
on the stone tools, although this overlap may be overestimated due to the lack of
universal starch classification types in the discipline.

These studies have provided some information on elements of Neanderthal

foraging strategy and show that Neanderthals were capable of sourcing nutrients

40
from a variety of plant foods. However, these findings only constitute qualitative
information from archaeological sites dispersed by space and time. More recent
work by Henry and colleagues (2014) attempted to explore Neanderthal plant use by
comparison to African Middle Stone Age and Near Eastern and Europe Upper
Palaeolithic peoples. This research used both dental calculus samples and wash
samples of the surfaces of stone tools from Neanderthals, African Middle and Later
Stone Age, and Eurasian Upper Palaeolithic peoples. A Poisson mixed model was
used to test if Middle Stone Age and Middle Palaeolithic groups used less plants
than Upper Palaeolithic and Later Stone Age peoples, using number of plant types
as a metric of diet breadth, and controlling for the effects of geographic region, stone
tool type and sites. This model suggested that all of the considered groups
consumed an approximately comparable array of plant foods and none of the
expected parameters of variation (stone tool industry or geographic region) had a
significant influence on the number of plant species consumed. There was also no
apparent pattern in plant use through time. Fundamentally, the results failed to
detect any difference between Neanderthals and any modern human group.

These studies have indicated the potential for dental calculus research to
reveal foods, in particular the use of low-ranked underrepresented food sources (e.g.
Triticeae) and details about the breadth of plants consumed. However, there are
many aspects of the calculus record that must be considered when applying this
method to Neanderthal samples. In the following two sections, I discuss issues faced
when interpreting the dietary signal of dental calculus.

2.6 The state of the art in dental calculus research

2.6.1 A background

The study dental calculus has a long history in archaeological research.

Archaeological dental calculus has been noted as a pathology in studies of health of
past populations since early decades of the 20th century (Leigh, 1925; Hughes, 1963).
It was long recognised that this pathology is intertwined with diet, and the incidence
of dental calculus was studied as a proxy for the amount of carbohydrates or
proteins ingested (See section 2.5.6). The potential of dental calculus to open a door
to specific dietary choices of past populations was first noticed in the 1970-1980s.
Dobney and Brothwell (1986; 1988) demonstrated that dental calculus could yield
data on the diets of human populations. Today, analysis of plant and animal

41
microremains recovered from archaeological dental calculus has grown to become a
widespread means of aiding dietary and health reconstruction (Boyadjian et al.,
2007; Blondiaux and Charlier, 2008; Henry et al., 2011; Liu, 2012; Mickleburgh and
Pagán-Jiménez, 2012; Warinner et al., 2014; Power et al., 2016). Microscopic plant
remains preserved in dental calculus can inform us about the exploitation of plants
otherwise invisible to us, thereby enabling us to obtain direct information on a wide
variety of question on prehistoric societies. For example, plant microremains from
dental calculus have indicated the use of beans in a complex plant diet in South
America (Piperno and Dillehay, 2008), described early agricultural diets at Tell al-
Raqā'i, Syria (Henry and Piperno, 2008), and recorded pre-Columbian Caribbean
subsistence (Mickleburgh and Pagán-Jiménez, 2012). Dental calculus has also
contributed to dietary studies of the early African hominin Australopithecus sediba.
Phytoliths found in the dental calculus of the MH2 individual suggested a C3 diet
incorporating dicotyledons (tree leaves, fruits, wood and bark) and monocotyledons
(grasses and sedges) (Henry et al., 2012). More recently, dental calculus has been
used to examine characteristics of diet from Lower Palaeolithic hominins at Qesem
Cave in Israel (420–200 ka). Hardy and colleagues (2015) used starch grains and
specific chemical compounds recovered from dental calculus to infer the ingestion of
plant foods. They interpreted pollen, fungal spores, microcharcoal and invertebrate
remains as evidence of the inhalation of respiratory irritants.

2.6.2 Technical difficulties in current approaches in dental calculus research

Despite this growing interest in dental calculus as a source of ancient dietary

information, dental calculus is still poorly understood. Dentistry research has paid
scant attention to the mechanisms by which plant microremains become trapped and
preserved within calculus. Native starch grains (i.e. starches in their original
unaltered state) are the predominant focus of much of the microbotanical
archaeology literature (Mickleburgh and Pagán-Jiménez, 2012; Leonard et al., 2015;
Tao et al., 2015). Yet there has been a lag in explaining how starch grains and grain
morphology persist in archaeological contexts (Haslam, 2004; Torrence and Barton,
2006; Hardy et al., 2009). Starch is a biodegradable molecule and it should rapidly
degrade after burial (Hardy et al., 2009; Langejans, 2010; Henry, 2014). Starch does
seem to survive in certain situations, as unambiguous ancient starch is found in
archaeological contexts (Samuel, 1996). The survival of ancient starch presumably
reflects protective qualities of its semi-crystalline polysaccharide structure and

42
specific microenvironment conditions that isolate starches from taphonomic
processes (Hardy et al., 2009; Salazar-García et al., 2013; Henry et al., 2014).
However, phytoliths are far more robust than are starches and routinely survive in
most sediment types, yet they may dissolve when exposed to a high pH. It is poorly
understood how the alkaline environment (pH ≤ 9) of dental calculus affects
phytoliths (Kleinberg, 1970). Other botanical microremains that are useful for
archaeobotanists have similar problems. Calcium oxalate crystals (calcium
phytoliths) are a category of microremains found in energy-rich plants.
Problematically, calcium oxalate crystals are susceptible to dissolution even in mild
acids. The acidity of saliva may readily drop low enough to dissolve calcium oxalate
present in the mouth (Tromp, 2012). The other types of microremains that may be
present in dental calculus such as pollen, may well be subject to comparable
problems.

Dental calculus could offer these microremains a secure mineralised matrix

where they become embedded and entirely isolated from soil chemistry and
microbial action (Warinner et al., 2014). Of course, to even reach this point, starches
that have entered the mouth must first survive breakdown in the oral cavity. The
mouth is a hostile environment for exposed starch grains because of the action of
salivary digestive enzymes and bacterial metabolic activity that will rapidly attack
and digest starches (Lukacs and Largaespada, 2006). Most digestion of starch occurs
at a later point in the digestive system due to the effects of pancreatic amylase, but
the high amounts of salivary amylase found in most human groups may still have an
impact. We may only speculate that some starch avoids oral enzymatic digestion and
is stochastically forced into protected niche areas of calculus. Alternatively, it could
be explained in a slower model, where starch (resistant starch, higher in amylose
content than typical starches) evades digestion and is gradually precipitated into
dental calculus (Hardy et al., 2009).

There has been little attempt to examine mechanisms that may be involved.
Regrettably, the conventional methodologies in dental calculus analysis rely on
invasive sampling of calculus from the tooth, making this harder to study. They
involve mechanically or chemically removing dental calculus from the enamel
surface, grinding or dissolving it to break up the sample, and finally examining the
particles using optical light microscopy (Henry and Piperno, 2008). Due to this
extraction, microremains observed by the analyst are no longer in context in
calculus. This is unfortunate, because the microenvironments that seem to preserve
microremains in dental calculus may shed light on whether microremains are not lab

43
contamination, the mechanisms of microremain preservation, and if they are
damaged in extraction. Due to this, Weyrich and colleagues have questioned
reproducibility and accuracy of dental calculus studies using microremains, pointed
to sedimentary contamination as undermining these studies (Weyrich et al., 2015).

2.6.3 The representativeness of the dental calculus dietary record

Perhaps the most serious problem in dental calculus dietary studies is the
ambiguity of what can be inferred about diet from the plant microremain record
found in dental calculus. Researchers can say surprisingly little about how
representative this record is, despite the plethora of studies using dental calculus as
a source of dietary information. Little research has attempted to quantitatively cross-
validate the dietary material recovered in dental calculus with the organism’s actual
feeding ecology and life history. Detailed studies using controlled diets have not
been pursued and there are few published studies using faunal dental calculus
where diet may be reliably predicted (Armitage, 1975) . Many past studies have
assumed that the plant matter preserved in dental calculus representatively reflects a
long-term dietary average (Mickleburgh and Pagán-Jiménez, 2012). Yet we should
question this assertion for many good reasons. Plant remains trapped in dental
calculus could plausibly be derived from airborne particles, water, chewing of plant
matter for health or fibre processing, amongst other possibilities. This raises the
prospect that microremains found in dental calculus from the studied Neanderthal
samples may not in fact reflect plant consumption at all.

Knowledge of the timing of the formation of the calculus dietary signal would
greatly assist life-history based approaches. Dental calculus does not form and
accumulate in a continuous and predicable way (Lieverse, 1999). Soft dental plaque
can make the transformation into hard mineralised calculus over the course of
weeks, but mineralization may be episodic and the rate it occurs varies among
individuals according to age, oral hygiene, nutrient intake (Bergström, 1999;
Lieverse, 1999; Jin and Yip, 2002) and possibly also genetic predisposition among
other things. In addition to these complications, dental calculus deposits can become
dislodged from the enamel at any point during life, resetting the dental calculus
dietary record along with it.

Leonard and colleagues examined dietary representativeness using living

Namibian Twe forager-horticulturists (2015). The Twe retain a partially traditional

44
diet in a mountainous arid habitat. Twe grow maize (Zea mays), pearl millet
(Pennisetum glaucum), squash (Cucurbita sp.), melons (Cucurbitaceae spp.) and
sugarcane (Saccharum sp.). Although they still collect a variety of wild plant foods,
since 2006 an increasing proportion of their diet is from government-supplied maize
meal. Leonard and colleagues established dietary patterns through interviewing and
observing food choice during in short-term camp stays. Leonard and colleagues
noted that older individuals and males had larger dental calculus deposits than
young people (under 35 years old) and females, a potentially confounding affect.
The number of microremains per individual poorly predicted the range of starch and
phytolith producing plants consumed. Nineteen Twe vegetal foods contained starch
and phytoliths, but no calculus sample contained more than six plant food types. A
population approach was more successful, but Leonard and colleagues stressed that
in her population, a sample of 50 individuals or more is needed to have 95 %
confidence of observing several foods. Overall, their analysis suggested that starch
grains and phytoliths in Twe dental calculus gave an incomplete picture of diet and
significantly underrepresented vegetal foods (Leonard et al., 2015). It is unclear if
these results can be applied to prehistoric hominin populations. It is unknown how
many copies of AMY1 the Twe people have so we cannot assess if this influenced the
results. Despite this study’s valuable contribution, this study lacked insight into the
long-term dietary history of the studied individuals.

2.6.4 Outlining the findings for dental calculus research and Neanderthal diet

Sections 2.1 through 2.3.2 provide details why reconstructing diet is an

essential part of studying both human origins and the life history of Neanderthals. A
behavioural ecology framework is a very powerful means for achieving this task. By
using diverse array of approaches, I have demonstrated what can be extrapolated
from current evidence about the Neanderthal diets. I have assessed the crucial issues
that must be resolved to move forward to a more complete palaeobiology of these
hominins. This dissertation is based on using plant consumption to test how
adaptable Neanderthal diets were. Dental calculus analysis was selected to provide a
window on the plant use of Neanderthals. The unanswered questions highlighted by
this introductory section form the basis of this PhD, which includes three main parts:
are part of this PhD; a revision of how we obtain dietary data from dental calculus, a
re-evaluation of the resolution of the dental calculus dietary record and a new
measure of plant foods and dietary breadth for Neanderthal diets.

45
 The first paper highlights the problems of conventional dental calculus
research. It examines dental calculus from wild chimpanzees and archaeological
specimens first with scanning electron microscopy–energy-dispersive X-ray
spectroscopy, and then with conventional optical light microscopy to compare
techniques. This allowed for the first time investigation of the microenvironment
that traps starch and other microremains. Lastly, it developed a sequential workflow
that maximises the amount of life history information extractable from dental
calculus. The second paper focuses on associating debris of plants inside dental
calculus to diet and behaviour. It aims to address the troubling gap between plants
in dental calculus and dietary records by using calculus of chimpanzees with
documented diets. Samples of chimpanzee dental calculus (Taï National Park, Côte
d’Ivoire) showed that microremains accumulate as long-lived dietary markers. The
paper found that phytoliths allow feeding preferences of the chimpanzees to be
reconstructed, while starches do not. Microremains also implied that assemblages
could record population information about other dietary behaviours, such as the age
of weaning and learned food processing techniques like nut cracking. Finally, the
third paper uses dental calculus from Neanderthal remains to provide new light on
plant exploitation from a mix of environments. Dental calculus was analysed from
five archaeological sites: Vindija (Croatia), Grotta Guattari (Italy), Grotta Fossellone
(Italy), Sima de las Palomas del Cabezo Gordo (Spain) and Kalamakia (Greece).
These sites represent a variety of regions and biomes across Europe. Starch,
phytoliths and other microremains suggested Neanderthals used a wide variety of
plants including low ranked plant foods. The findings were then combined with
data from past studies to model if local vegetation, winter temperature or the age of
the site account for variation in diet. The model found local vegetation and winter
temperatures do not influence the patterns in the dental calculus data suggesting
that although Neanderthal consumed they have had an inflexible but partially broad
dietary adaptation.

46
 3

Assessing use and suitability of scanning electron

microscopy in the analysis of microremains in dental
calculus

Robert C. Power,1 Domingo C. Salazar-García,1,2 Roman M. Wittig,3 Amanda G. Henry1

1Max Planck Research Group on Plant Foods in Hominin Dietary Ecology, Max Planck Institute for
Evolutionary Anthropology, Deutscher Platz 6, 04103 Leipzig, Germany.
2Department of Human Evolution, Max-Planck Institute for Evolutionary Anthropology, Deutscher

Platz 6, Leipzig, Germany.

3Department of Primatology, Max Planck Institute for Evolutionary Anthropology, Deutscher Platz 6,
Leipzig, Germany.

Published in the Journal of Archaeological Science (2014, Volume 49, 160-169).

Abstract

Dental calculus is increasingly recognised as a major reservoir of dietary

information. Palaeodietary studies using plant and animal microremains (e.g.
phytoliths, pollen, sponge spicules, and starch grains) trapped in calculus have the
potential to revise our knowledge of the dietary role of plants in past populations.
The conventional methods used to isolate and identify these microremains rely on
removing them from their microenvironment in the calculus, thus the
microenvironment that traps and preserves microremains is not understood. By
using scanning electron microscopy and energy-dispersive X-ray spectroscopy
(SEM-EDX) on modern chimpanzee calculus from the Taï Forest, Côte d'Ivoire, and
human calculus from the Chalcolithic site of Camino del Molino, Spain, we present
the first reported observations on characteristics of the matrix setting that are
conducive to the survival of starch in dental calculus. We also assess the potential for
SEM-EDX to detect starch and differentiate it from structurally and molecularly
similar substrates. We demonstrate that SEM-EDX may offer a non-destructive
technique for studying microremains in certain contexts. Finally, we compare
traditional optical analytical techniques (OM) with less invasive electron
microscopy. The results indicate that SEM-EDX and OM are both effective for

47
observing microremains in calculus, but differ in their analytical resolution to
identify different microremains, and we therefore recommend a sequential use of
both techniques.

3.1 Introduction

Dental calculus, or dental plaque calcified by salivary calcium phosphate, was

first noticed as a reservoir of dietary information when Armitage (1975) recognised
plant remains on the teeth of archaeological ungulates. Dobney and Brothwell (1986,
1988) later demonstrated the value of calculus in the study of human diets. Analysis
of plant and animal microremains in archaeological dental calculus is a rapidly
growing field in dietary reconstruction (e.g. Boyadjian et al., 2007; Henry et al., 2012;
Liu, 2012; Mickleburgh and Pagán-Jiménez, 2012; Warinner et al., 2014). Researchers
have reported starch, phytoliths, pollen, diatoms, chrysophycean cysts, sponge
spicules, and mineral particles in human calculus up to tens of thousands of years
old (e.g. Dobney and Brothwell, 1988; Boyadjian, 2012).

Despite this interest in dental calculus as a source of dietary information,

there are still many questions about the mechanisms by which plant microremains,
particularly starch grains, are preserved within the calculus. Native starch grains (i.e.
starches in their original, unaltered state) are the major focus of many recent and
ancient dietary studies. Starch is a foremost nutritional component in many human
and non-human primate diets, and it can also survive in the archaeological record
over long periods of time due to its semi-crystalline polysaccharide structure
(Mercader et al., 2008; Hardy et al., 2009; Henry et al., 2011; Salazar-García et al.,
2013; Leonard et al., 2015). The means by which starch embeds and preserves in
calculus is still unclear. The mouth is a hostile environment for starch preservation
because of the action of salivary digestive enzymes and bacterial metabolic activity
(Lukacs and Largaespada, 2006). Calculus forms gradually as bacteria-rich plaque
biofilms mineralise from calcium phosphate in the saliva over a period of days to
years (Abraham et al., 2005). During this formation and mineralization process, the
starch grains are exposed to α-amylase, which is present in the saliva of humans and
several orders of mammals (Butterworth et al., 2011). Amylase quickly digests starch
by breaking down the polysaccharide crystalline structure into various simple and
complex sugars through hydrolysis (Lukacs and Largaespada, 2006). Theoretically,

48
starch may avoid oral digestion and survive in protected niche areas in calculus, but
this has not yet been empirically confirmed.

In addition to the difficulties with starch preservation in the oral cavity, there
is also the possibility that the starches that have been recovered from calculus are
actually the result of modern contamination. Modern starches are abundant in the
air, water, and working surfaces of most facilities, making environmental
contamination a strong possibility. Archaeological and field site contexts suffer from
sources of contamination such as airborne starch rain, but the greatest risk of
contamination comes from excavation and post-excavation handling in the presence
of food or due to the use of gloves powdered with corn or other starches (Newsom
and Shaw, 1997; Loy and Barton, 2006; Laurence et al., 2011).

Currently, the standard methodology for starch grain recovery from calculus
is too destructive to confirm whether observed starch came from the calculus or
from contamination. This method involves mechanically or chemically removing
calculus from the tooth, grinding or dissolution to break up the sample, and finally
examining the particles using optical light microscopy (OM) (Henry and Piperno,
2008). Furthermore, to the untrained eye, several other calculus components, such as
cysts, mineral grains, fungal spores, wood cells, and air bubbles may be confused
with starch grains when viewed only under OM. Some have proposed confirming
starch presence by measuring amylase activity on treated samples (Hardy et al.,
2009), but this enzyme destroys the starch in the process. One common and reliable
means to detect starch is to apply iodine potassium iodide (IKI) solution, which
binds to the amylose molecule, and look for the characteristic blue-black stain.
However, this temporarily obscures the starch’s diagnostic surface features.
Furthermore, it is impractical to apply a staining solution to an intact calculus matrix
because objects within the mineralised matrix are protected from moisture.
Accordingly, there is a great need for more sophisticated and non-destructive
methods to confirm the successful detection of starch grains in dental calculus. Some
researchers have suggested the possibility of using scanning electron microscopy
(SEM) to study plant microremains in calculus (Dobney and Brothwell, 1986;
Reinhard et al., 2001; Tromp, 2012). Despite the success of this method in locating
phytoliths (Arensburg, 1996; Lalueza-Fox et al., 1996; Charlier et al., 2010; Kucera et
al., 2011; Tromp, 2012; Tao et al., 2015), the detection of starch grains through SEM
has not yet been attempted.

49
 In this study, we present SEM coupled with energy-dispersive X-ray
spectroscopy (EDX) as a novel means for identifying starch and other microremains
in intact human and chimpanzee dental calculus. This system provides us with the
ability to identify microremains, including starch grains, by their morphology and
elemental composition in situ in the calculus, thus ruling out contamination. It also
allows us to explore the kinds of environments within the calculus that may permit
starch preservation. Furthermore, we examine the potential of EDX to detect starch
by comparing the elemental makeup of native starch to those of saliva-hydrolysed
starch and other non-starch saccharides to learn whether EDX distinguishes starch
from other polymers of similar elemental makeup. This identification allows us to
positively show that starch grains survive in calculus. Finally, we compare the
results from SEM-EDX to those from OM on the same human calculus samples to
determine whether these techniques offer comparable or complementary results.
Due to time constraints, we were unable to conduct this portion of the analysis on
the chimpanzee samples and instead only used human dental calculus samples.

3.2. Materials and methods

3.2.1 Study groups

The calculus samples were obtained from two groups, modern wild
chimpanzees from the Taï Forest in Côte d'Ivoire and humans from the Chalcolithic
collective burial of Camino del Molino in Spain (Table 4). We chose these two test
groups for the following reasons: 1) individuals from both have abundant calculus
on their teeth; 2) they represent modern (chimpanzee) and archaeological
(Chalcolithic humans) timeframes; and 3) both groups maintained very different
dietary strategies and should therefore have different microremain profiles.

The sample of chimpanzee calculus came from the Taï Chimpanzee Osteology
Collection curated at the Max Planck Institute for Evolutionary Anthropology (MPI-
EVA) in Leipzig, Germany. The behaviour of the wild chimpanzees living in the Taï
Forest has been monitored and documented since the commencement of the Taï
Chimpanzee Project in 1979 (Boesch and Boesch-Achermann, 2000). Taï Forest data
collection complied with the requirements and guidelines of the Ministère de
l’Enseignement Supérieure et de la Recherche Scientifique, and adhered to the legal
requirements of Côte d’Ivoire. The osteology collection contains 77 chimpanzees. We
chose calculus samples from individuals who had comprehensive observational

50
records documenting diet, sex and age. After their death, the remains of these
individuals were interred for defleshing, later exhumed, and curated. We collected
calculus from molars or canines of six individuals; two females and four males. The
Taï Chimpanzees consume native starch from wild nuts and seeds such as the Gabon
nut (Coula edulis Baill.) and Kola nut (Cola nitida (Vent) Schott et Endl.) (Hohmann et
al., 2010; N’guessan, 2012), and unlike humans, they consume no cooked or
processed foods. Our preliminary reference collection of Taï Forest chimpanzee
foods shows that ten of the 82 foods we analysed are starch-rich. However, these 82
species represent less than a third of plants this population is known to consume,
and we are still building this reference collection. Chimpanzees also produce
salivary amylase, though purported at much lower quantities than do humans
(Perry et al., 2007; Behringer et al., 2013).

Camino del Molino is a Chalcolithic collective burial pit found during

construction work in the city of Caravaca de la Cruz (Murcia, southeast Spain).
Radiocarbon dates from bone collagen samples spanning the burial sequence
indicate that the site was in continual use over a span of 300- 400 years during the
first half of the third millennium B.C. The site contained a minimum of 1,300
individuals, likely the remains of 16-20 generations of one population buried at one
place (Lomba Maurandi et al., 2009). Approximately 30 % of the individuals are
classified as juvenile (<14 yrs.), and the rest are adults spanning from young to old
(Haber Uriarte et al., 2013). We collected dental calculus preferably from lower
molars for standardisation from the teeth of six individuals; two females, two males,
and two individual of unknown sex (Table 4). There are no archaeobotanical studies
from Camino del Molino or from the broader region of Murcia contemporary to the
site. However, studies of Late Neolithic and Chalcolithic deposits in neighbouring
regions suggest that the number of cultivated species is low and consists mainly of
naked wheat (Triticum sp.), barley (Hordeum vulgare L.), some lentil (Lens culinaris
Medikus) and common vetch (Vicia sativa L.) (Pérez Jordà, 2005; Pérez Jordà and
Carrión Marco, 2011). There is no published study from the site on culinary
practices, in part because it is a necropolis and not a habitation site. Despite this, its
Chalcolithic age indicates that this population consumed cooked food, because
cooking is widespread across the European Neolithic and Bronze Age societies
(Thissen et al., 2010; Halstead, 2012).

51
Table 4: Calculus samples analysed using SEM-EDX and OM. Sex and age classification of the Camino del Molino remains
are preliminary (Haber Uriarte et al., 2013).

Lab identifier Individual number Type Tooth Sex Age (years) Weight
SJ-13-32 Sujeto 6 Camino del Molino P1 mandible F 26-28 1.76 mg
SJ-13-33 Sujeto 8 Camino del Molino C maxilla M 22-24 0.51 mg
SJ-13-36 Sujeto 11 Camino del Molino I2 mandible ? ? 6.06 mg
SJ-13-37 Sujeto 17 Camino del Molino M2 mandible M? 43-55 10.0 mg
SJ-13-38 Sujeto 113 Camino del Molino M2 mandible F 24-28 1.88 mg
SJ-13-39 Sujeto 151 Camino del Molino C mandible M 30-35 1.09 mg
Venus 15001 Taï Chimpanzee M3 maxilla F 27 0.72 mg
Leo 15012 Taï Chimpanzee M3 mandible M 19 1.19 mg
Fanny 11780 Taï Chimpanzee M3 mandible F 25 3.34 mg
Goma 15004 Taï Chimpanzee M3 mandible F 28 2.40 mg
Rubra 15023 Taï Chimpanzee M3 mandible F 38? 3.88 mg
Castor 13439 Taï Chimpanzee M3 mandible F 22 2.25 mg

3.2.2 Calculus sampling

We selected teeth encrusted with a prominent band of calculus present on the

enamel surface. We sampled only supragingival calculus (above the gum line), since
it is unclear if subgingival calculus (below the gum line, on the neck of the tooth)
preserves food remains. We photographed the calculus before sampling, and then
brushed the sample tooth gently with a dry, sterile toothbrush to remove surface
contaminants. We then used a dental scalar to remove small areas of supragingival
calculus (~4 mm area), from the enamel. We conducted all calculus sampling in a
positive pressure hood at the archaeological science laboratories at the MPI-EVA. We
then weighed each of the samples and transferred them to microcentrifuge tubes for
storage until further use. Following sampling, the teeth and surviving calculus were
photographed again. Additionally, we collected control samples, including the
packing material in which the teeth had been stored.

3.2.3 Electron microscopy analysis

We conducted the SEM-EDX analysis at University College Dublin’s Nano-

Imaging and Materials Analysis Centre (NIMAC) in Dublin, Ireland. The calculus
samples were mounted on stubs using double-sided carbon tape, and sputter coated
with gold for 20 seconds using an Emitech K575X Sputter Coating Unit, to prevent
surface charging by the electron beam. We then examined the calculus using a FEI
Quanta 3D FEG DualBeam (FEI Ltd, Hillsboro, USA) SEM with an attached EDAX
ED APOLLO XV Silicon Drift Detector with a 5 – 10 kV accelerating voltage. EDX

52
detected and documented most elements of interest excluding hydrogen, which is
non-detectable with this method. We omitted the gold elemental peak from each
spectrum since the gold was added during sputter coating. We photographed and
documented every tentative microremain and later described our observations.

3.2.4 Carbohydrate reference standards and partially hydrolysed controls

We used a variety of reference standards (see Table 5) to assess the accuracy

of EDX reads on the experimental sample types of starch. Starches from a variety of
plants were selected to represent major starch types such as corn starch, potato
starch, and common dietary components for each population (Boesch and Boesch
1983): wheat (Triticum aestivum L.), Gabon nut (Coula edulis Baill.), Xylia (Xylia evansii
Hutch.) and Kola nut (Cola nitida [Vent] Schott et Endl.). The nuts were ground,
dried and weighed to derive nut flour suitable for use. Wheat, potato and corn were
purchased from local distributers in Germany (see Table 5).

Table 5: List of reference samples analysed using EDX.

Reference sample Part Type Source

Fructose N/A Lab-grade Roth - 4981.1
Sucrose N/A Lab-grade Roth - 4621.1
Maltose N/A Lab-grade Roth - 8951.1
Glucose N/A Lab-grade Sigma - G7528
Maize (Zea mays subsp. Mays L.) Grain Cornstarch Speisestärke, RUF Lebensmittelwerk
Kola nut (Cola nitida (Vent) Schott et Endl.) Nut Bulk plant Collected in Taï National Park
Xylia (Xylia evansii Hutch.) Nut Bulk plant Collected in Taï National Park
Gabon nut (Coula edulis Baill.) Nut Bulk plant Collected in Taï National Park
Potato (Solanum tuberosum L.) Tuber Flour Kartoffelmehl, RUF Lebensmittelwerk KG
Wheat (Triticum aestivum L.) Grain wheatstarch Weizella, Hermann Kröner GmbH

Laboratory-grade fructose, maltose, glucose and sucrose (Roth, Germany)

were included as standards because they have nearly identical elemental
compositions as starch but with structurally different molecular arrangements (e.g.
sucrose has 2.1 wt % (mass fraction) more carbon than fructose, but 2.1 wt % less
than starch).

To compare EDX element signatures for the different types of saccharides, we

took EDX measurements from five individual grains of fructose, sucrose, maltose,
glucose and wheat starch, corn starch, Kola nut starch, Xylia starch and potato
starch. This allows the comparison of a monosaccharide, a disaccharide, and a
polysaccharide (starch).

53
 Finally, to assess whether EDX signatures and detection accuracy is affected
by the salivary modification (hydrolysis) of starch, we experimentally hydrolysed
the native starches from the wheat flour and both nut varieties using salivary
amylase derived from human saliva – a simulation of the effects of oral digestion on
starch that can occur. One of us (R.C.P.) provided the saliva used in all experiments,
which was collected on a single occasion. We split each of the individual plant
samples into nine subsamples of approximately 2 mg each: three subsample per
plant remained untreated (control), three were exposed to amylase (35 µL of saliva)
for 30 minutes, and three were exposed to amylase (35 µL of saliva) for 90 minutes.
We also similarly partitioned the wheat flour into nine aliquots into three
subsamples of 2 mg each for identical amylase treatment. We ceased hydrolysis by
displacing the saliva with alcohol and centrifugation at 1691 x g to remove as much
liquid from the sample as possible and stop hydrolysis. Then the remaining alcohol
was evaporated at 35 º C in a drying oven. We performed measurements using SEM-
EDX in triplicate on one starch grain from each subsample, creating nine readings
per category (e.g. wheat 30 minute hydrolysed). A summary of these analyses is
provided in Figs. 3 and 4.

3.2.5 Optical microscopy analysis

We performed optical microscopy on the ancient human remains at the Plants

Working Group Laboratory in the MPI-EVA, Leipzig. We removed the gold plated
calculus samples from the SEM mounting stubs, and then ground them in a 1.5 ml
Eppendorf microcentrifuge tube with a micro pestle containing ~50 µl of a 25 %
glycerine solution to reduce sample loss due to static electricity. Glycerine was
chosen as its refractive index is lower than starch making it suitable for starch
detection. The samples were then centrifuged at 1691 x g (Heraeus MEGAFUGE 16
with TX-400 Swinging Bucket Rotors) for 10 minutes. All of the resulting pellets
were mounted on glass slides and examined under bright field and cross-polarised
light on an A1 Zeiss Axioscope microscope at 400 × magnification. Larger samples
were mounted on several slides. Each microremain was photographed and
described (Table 6).

54
 3.3. Results

3.3.1 Standards

The EDX spectrum of starch is distinct from other saccharides but not
sufficiently to permit reliable identification (Fig. 3, Fig. 4; Appendix table 1,
Appendix table 2). The EDX results from all the samples indicate that oxygen is
underrepresented. Though carbon comprises roughly 40-50 wt % of these
saccharides, the EDX spectra indicates carbon at 60-90 wt % (Fig. 3). Comparing the
short-chain saccharides to the starches there is a difference but some types of starch
overlaps with each short-chain saccharides. This indicates that some starch may be
distinguishable from short-chain saccharides through EDX. There was far more
variability in carbon values in starch than in short-chain saccharides. Starch is
composed of oxygen, hydrogen and carbon (C6H10O5)n, where n ranges from 300 to
1000, so starch is approximately 42.1 %=carbon, 6.5 %=hydrogen, 51.4 %=oxygen
(Newman et al., 1996). Thus, maize starch comes closest to the expected values of
starch. This variability possibly reflects the heterogeneous nature of the starch.
Starch varies in both proportion of amylose and amylopectin and minor compounds
such as proteins and lipids (Belitz et al., 2009). We see further evidence of this
elemental variability in the native starch samples in Fig. 4, which had a higher
variability of both carbon and oxygen than the hydrolysed starches. The EDX
profiles of hydrolysed starches fall within the range of their native counterparts, yet
they show noticeable less variation and reduced oxygen values (Fig. 4). The
reduction in variation and lower oxygen levels in these samples may be either from
the result of the added ethanol reducing oxygen in the starch or the ethanol washed
off debris on the starch surface. A few of the damaged starches have slightly
increased oxygen percentages, but this is not consistent across all hydrolysed
subsamples. We found no evidence that saliva-activated hydrolysis could obscure
the starch EDX elemental signature. However, when large starch shaped objects are
present under SEM, it is possible to test whether these particles have a molecular
make-up that is similar to starch and other saccharides.

55
Fig. 3: Carbon wt % (mass fraction) of starch, sugars produced by hydrolysis and reference sugars. Plot shows
five individual grains of starches, glucose, and maltose and out-group carbohydrates (fructose and sucrose)
detected with EDX. Values exclude contaminating elements such as potassium from sweat (3.6.1).

Fig. 4: Comparing native starch versus samples that were hydrolysed with amylase for 30 and 90 minutes at
room temperature. Three starches were sampled with triplicate readings. Values exclude contaminating
elements such as potassium such as potassium from sweat (3.6.2).

56
 3.3.2 Calculus samples

Examination of the SEM images of the calculus confirmed that this matrix has
a heterogeneous texture, with smooth surfaces as well as many pores, crack and
crevices (Fig. 5, Fig. 6). Most of the pores appear to be the result of rod bacterial
pseudomorphs, which are shallow and measure only between 0.3 -1 µm in width
(bottom left in Fig. 5, and widely scattered in Fig. 6). These pores are too small to
preserve microremains, but larger cracks and crevices had many microremains (Fig.
6). Further examination of the calculus revealed several types of inclusions within
the matrix. In some cases, these inclusions were consistent with the overall size (15-
40 μm) and shape (ovoid- pyriform) of certain starch grain types, and inconsistent
with other microremains such as yeast and bacterial cells (Fig. 5). The supposed
starch clusters were clearly embedded in the matrix, with grains occluded by
overlying deposits of the matrix material. Interestingly, the starch grains were not
evenly distributed in the matrix, and often appeared in clumps (Fig. 5). This could be
explained in two ways; i) plant microremains are deposited in groups originating
from clumps in food lumps, or ii) microremains are only preserved in localised
niches, such as larger cracks and crevices, in the calculus matrix.

57
Fig. 5: Starch grains located in-situ on dental calculus surface. SEM image showing a group of starches
trapped in the matrix of one of the chimpanzee dental calculus samples (Venus), with the corresponding EDX
spectrum (right) showing a calcium phosphate and silicon mantle covering a carbon rich starch (A) and solely
a carbon-rich starch (B).

58
Fig. 6: SEM image showing microremain diversity. A concentration of pollen (A) and sponge spicules (B) in
SJ-13-39 from Camino del Molino. Microremains were often found clustered.

Fig. 7: SEM image showing localised damage that arises from higher primary voltage SEM (10 kV) and EDX
on a spicule in calculus from Camino del Molino. Before (left); after (right).

59
 The EDX spectra of the calculus matrix from all of our samples indicate that it
is mostly composed of calcium and phosphorus, with trace amounts of aluminium,
magnesium, silicon, sodium and manganese (Appendix table 3). These elements
confirm our supposition that the majority of our samples consist of calculus, a
mixture of hydroxyapatite and other minerals, rather than contaminating exogenous
matter (Charlier et al., 2010; Salazar-García et al., 2014). In some instances, silicon
was locally abundant in the calculus (Appendix table 3), which may be important for
the preservation of starch grains. In contrast to the mineral matrix, the suspected
starch clusters, such as on chimpanzees Venus and Castor had significant carbon
peaks. Additionally, the starches often had calcium and phosphorus peaks,
reinforcing visual observations that they were indeed embedded in calculus (Fig. 5).
The combination of shape and elemental data (Fig. 5) is strongly suggestive of in-situ
findings of microremains preserved in the dental calculus environment. This is
possible as starch is morphologically distinct from other carbon rich particles such as
fungal filaments, Candida albicans cells, cellulose and sugars. We also note that the
starch we located with SEM-EDX was undamaged and we did not locate any semi
gelatinised or hydrolysed starch. These were also not observed with optical
microscopy but it is possible damaged starch was not visible with this approach.
In addition to the starches, we also identified a variety of other plant and animal
microremains preserved in the calculus using SEM-EDX, including phytoliths,
sponge spicules, diatoms and pollen (Table 6). These microremains were identified
by their diagnostic morphology using conventional methods (Torrence and Barton,
2006; Nadel et al., 2013; Power et al., 2014a), and this identification was confirmed by
their EDX spectra. For example, spicules were easily identified based on their long
rectangular shape and high level of regularity (Fig. 6 and Fig. 7) unlike smooth long-
cell phytoliths, and EDX readings confirmed their biogenic silica composition (7.1).
OM also demonstrated the presence of a rich assemblage of plant microremains
(Table 5). We noted some of these microremains during the SEM analysis, such as
the abundant monaxon spicules (Fig. 6), but we only detected some, such as multi-
cell long-cell phytoliths, unsilicified plant cells and calcium oxalate (Fig. 8), with
OM.

60
 Table 6: Recovered microremains using both microscopy approaches.

Scanning electron microscopy Optical microscopy

Taï chimpanzees Camino del Molino Camino del Molino

SJ-13-32
SJ-13-33
SJ-13-36
SJ-13-37

SJ-13-38

SJ-13-39
SJ-13-32

SJ-13-33

SJ-13-36

SJ-13-37

SJ-13-38

SJ-13-39
Castor
Venus

Fanny

Rubra
Goma
Leo
Starch 29 2 3 4 40 22 3 1 1 6 1 8 10 1 3
Single-cell long- 1 1 1 1 2 3 5 10 1 3
cell
Multi-cellular 1 11 1
Long-cell
Short-cell 1 3 1 2 3
Phytoliths

Parallelepipedal 1 1 1 1 6 1
Bulliform 1 2
Plate 1 1 1
Rugulose 2 1
Spheroid
Smooth spheroid 3 2 1 1 1
Hair 2 1 1 1 1
Unidentified 1 1 3 2 3
Unsilicified plant cell 15
Prism calcium oxalate 5 8 2
Annular ring 2
Monaxon spicule 30 1 5 1 15 46 8 5 14 18 11 10
Quartz grain 1 2
Pennate diatom 20
Indet diatom 2 2
Echinate pollen 1 1 1 3 3
Other pollen 3 1 1 3 1 1 2
Chrysophycean cyst 4
Fungal filament + +
Fibre + 1
Invertebrate 1
Other 2
+ : a high but unquantified number

61
Fig. 8: A calcium oxalate prism observed with optical microscopy in SJ-13-37; under bright field optical
microscopy (left), and cross-polarising optical microscopy (right).

A comparison of the microremains observed under SEM-EDX with those seen in OM

revealed important differences (Table 6). We observed more starch microremains
using OM than SEM-EDX. This is probably because the sample preparation for OM
breaks down the calculus matrix, freeing starch microremains that were trapped in
the middle of the calculus chunk. Yet paradoxically, other microremains, such as
sponge spicules, were more commonly seen in SEM-EDX than in OM of the same
samples.

Based solely on the SEM results (we did not perform OM on the chimpanzee
samples), the two groups we studied did present some differences. The chimpanzee
samples were rich in starch grains and diatoms, while the human samples had an
abundance of unsilicified plant cells and sponge spicules (Table 6).

3.4. Discussion

Analysis of calculus samples by SEM-EDX and OM provides data that

validates the study of microremains recovered from this biological material. By
SEM-EDX, we were able to identify the elemental constituents of starch, and confirm
its position in situ in calculus particles. This is the first time that starch has been
identified by its elemental signature while still embedded within the calculus matrix,

62
and confirms that starch can be preserved in calculus, and can therefore be a useful
source of dietary information.

The analysis suggests that certain features of the calculus may promote the
preservation of microremains, and starch grains in particular. Food debris may trap
around calculus rather than calculus growing around food debris. While the pores
left from bacteria colonies were too small to provide a protected niche for starches,
larger cracks and crevices were full of microremains, possibly because these areas
provided a protected environment. Furthermore, the silicon we detected in the
dental calculus may be significant. Silicic acid can induce spontaneous precipitation
of calcium phosphate in the saliva, which is the precursor mineral necessary for
calculus formation. Silicic acid may be consumed directly via water or indirectly via
plants, as it enters plants along with groundwater. Consuming polysilicic acid and
silica increases calculus formation, thereby regulating this process (Damen and Ten
Cate, 1989; Roberts-Harry and Clerehugh, 2000; Jin and Yip, 2002). Our observations
of silicon concentrations adjacent to embedded starch clusters (Fig. 5) corroborates
these reports, suggesting that dietary exposure to silica or silicic acid enables
enhanced calculus formation and thus the preservation of native starch in dental
calculus.

By following the SEM analysis with an OM examination of the same samples,

we are able to compare the effectiveness of each for specific microremain types.
Sponge spicules were easily visualised under SEM, but were seen less with OM. This
may be because the spicules are relatively fragile and are damaged when the
calculus is processed, possibly explaining why spicules are rarely reported in dental
calculus studies (Tromp, 2012; Dudgeon and Tromp, 2014). Because these particles,
as well as diatoms and Chrysophycean cysts, are highly dependent upon water
sources, they may indicate source type and provenance of consumed water, making
them powerful potential ecological markers for primatology and archaeology studies
(Dudgeon and Tromp, 2014). In contrast, calcium oxalate crystals were only visible
under OM, and not SEM. These crystals, which may occur as druses, raphides or
other similar forms, are a potentially useful marker of plants. They may be more
visible using OM because they have high interference colours that are visible under
cross polarised light (Fig. 8). For reasons that remain unclear, calcium oxalate is
rarely reported or discussed in calculus literature. Some research indicates that
calcium oxalate does not survive due to acidity in the mouth (Tromp, 2012), but
given their sheer abundance in plants and the relatively neutral oral pH, it is likely
that calcium oxalates do survive and are simply overlooked. On the other hand,

63
starch grains were clearly visible using both SEM and OM. However, we did note
that within individuals, the starches that we observed under OM typically did not
match the size and morphology of those seen in SEM-EDX. This contrasts with the
spicules, which often matched size and shape. This is likely due to the small number
of starches but high number of spicules. We also observed pollen grains embedded
in the calculus using SEM (Fig. 6) and OM. Although this type of pollen grain were
too small to analyse with the EDX, we do believe that SEM-EDX may be appropriate
for identifying many larger types of pollen grains, since these plant remains are
composed of potassium, magnesium, sodium and calcium (Szczęsna, 2007) and
should be easily visible in the EDX spectra.

Finally, the SEM analysis accurately reflected some stark differences between
our study groups. The differences in microremains number and types between the
chimpanzee and humans likely reflect the dietary behaviour and the age of the
remains. The chimpanzees consumed only raw plants, while the human group
potentially cooked much of their food. The chimpanzees therefore consumed many
more native, undamaged starch grains, and so there is greater opportunity for the
preservation of native starch grains in dental calculus. Though the humans may
have consumed more starch overall, many of these starches would have been semi
gelatinised through cooking, disrupting the semi-crystalline structure and reducing
the potential for starch preservation in the mouth (Holm et al., 1988). Cooking,
combined with higher levels of salivary amylase in humans relative to chimpanzees
(Perry et al., 2007; Behringer et al., 2013) may have greatly reduced the relative
proportion of starch entering the human calculus matrix during its formation.
Furthermore, the chimpanzee samples are modern and likely to be well-preserved
while starch in the human calculus may have depleted due to digenesis over
thousands of years.

Overall, SEM-EDX does allow us to visualise and identify microremains

embedded in dental calculus, but this technique is not without limitations and
constraints. Internal features of starch grains that are vital for identifying the
taxonomic origin of the starch are not visible under SEM. We found that when using
EDX combined with higher primary voltage (10 kV), the beam moved or damaged
fragile microremains such as spicules (Fig. 7). EDX can only give reliable data on
objects ≥4 μm due to the penetration of the beam, making it impossible to measure
very small microremains including smaller starches. We found other techniques
such as backscatter detection to be of little additional advantage in detecting starch,
though this method may be useful in certain contexts such as examining calculus for

64
embedded phytoliths (Tromp, 2012). It is possible to examine only the surface
portion of intact calculus matrix using SEM-EDX, and so this is not a viable method
for visualising interior dental calculus structure and microremains. Sample
preparation may also be destructive since samples must be gold-plated and
mounted, but use of SEM without the plating may cause the sample image quality
and identification power to deteriorate.

3.5. Conclusions

The visual identification and subsequent elemental testing of microremains

embedded in the dental calculus of humans and chimpanzees suggests that these
important dietary markers are indeed trapped and preserved in calculus during the
lifetime of the individual. Clearly, this matrix has a protective quality that shields
fragile and degradable components, namely starch, from the enzymatic oral
environment.

SEM-EDX and OM have different sensitivity to different microremains. SEM-

EDX offers a means to confirm the presence of starch by combining morphological
and elemental information without having to destroy either the calculus, as required
in processing for OM, or the starch grains themselves, as proposed when using
enzymatic reactions. Even if starch is semi gelatinised it should preserve an
elemental signature that is suggestive of starch. We applied SEM-EDX to intact
calculus to witness microremains in situ, but this technique is equally viable for more
finely processed calculus samples mounted on plates, or even to calculus still
attached on the original tooth. However, it is important to note that diagnostic
features of starch grains, such as the hilum and lamellae, are only visible using OM.

Our study indicates that SEM-EDX is a viable alternative to OM analysis of

calculus, but researchers should choose their analytical method based on the
questions they seek to answer, and the plant microremains that they intend to study.
Furthermore, on very sensitive osteological remains, it may be possible to use SEM-
EDX to study calculus using entirely non-destructive means to examine embedded
microremains directly on the tooth; a useful technique if the tooth is not firmly
attached in the mandible or maxilla. We prefer to consider SEM-EDX a
complementary rather than replacement technique in the study of dental calculus
microremains. A sequential workflow that first examines calculus under SEM-EDX
and then under OM may be the optimal solution for highest resolution of
microremains, though we recognise that this approach is time intensive and can be

65
costly. We believe that further exploration and experimentation of SEM techniques is
important in the field of archaeological and palaeodietary reconstruction. The
continued refinement and expansion of dental calculus analysis techniques is an
important focus in order to optimise the information we can harvest from this
ephemeral and fragile material.

66
 4

Dental calculus evidence of Taï Forest Chimpanzee

plant consumption and life history transitions

Robert C. Power,1 Domingo C. Salazar-García,2,3,4 Roman M. Wittig,5,6 Martin Freiberg,7

Amanda G. Henry1

1Max Planck Research Group on Plant Foods in Hominin Dietary Ecology, Max Planck Institute for
Evolutionary Anthropology, Deutscher Platz 6, 04103 Leipzig, Germany.
2Department of Archaeology, University of Cape Town, Cape Town, South Africa.
3Departament de Prehistòria y Arqueologia, Universitat de València, València, Spain.
4Department of Human Evolution, Max Planck Institute for Evolutionary Anthropology, Deutscher
Platz 6, Leipzig, Germany.
5Department of Primatology, Max Planck Institute for Evolutionary Anthropology, Deutscher Platz 6
Leipzig, Germany.
6Taï Chimpanzee Project, Centre Suisse de Recherches Scientifiques, Abidjan, Cote d'Ivoire
7Institute of Botany, University of Leipzig, 04103 Leipzig, Germany.

Published in Scientific Reports (2015, Volume 5, 15161).

Abstract

Dental calculus (calcified dental plaque) is a source of multiple types of data

on life history. Recent research has targeted the plant microremains preserved in this
mineralised deposit as a source of dietary and health information for recent and past
populations. However, it is unclear to what extent we can interpret behaviour from
microremains. Few studies to date have directly compared the microremain record
from dental calculus to dietary records and none with long-term observation dietary
records thus limiting how we can interpret diet, food acquisition and behaviour.
Here we present a high-resolution analysis of calculus microremains from wild
chimpanzees (Pan troglodytes verus) of Taï National Park, Côte d’Ivoire. We test
microremain assemblages against more than two decades of field behavioural
observations to establish the ability of calculus to capture the composition of diet.
Our results show that some microremain classes accumulate as long-lived dietary
markers. Phytolith abundance in calculus can reflect the proportions of plants in the
diet, yet this pattern is not true for starches. We also report microremains can record

67
information about other dietary behaviours, such as the age of weaning and learned
food processing techniques like nut-cracking.

4.1 Introduction

Understanding feeding ecology is crucial for recognising the evolutionary

pressures that shaped the great apes and humans. It is long recognised that factors
such as dietary specialization, tool-assisted food acquisition and the weaning age of
infants are important in great apes and humans, and differ significantly among
species (Boesch et al., 1994; Ross, 1998; Boesch and Boesch-Achermann, 2000;
Teaford and Ungar, 2000).

However, many approaches to dietary reconstruction leave unanswered

specific questions on diet and related life history events, especially for fossil
specimens. There is a need for new methods to reconstruct food acquisition from
populations that can avoid some of the shortfalls of other techniques like direct
observation, stable isotope analysis and microwear studies (Phillips and Lancelotti,
2014; Fiorenza et al., 2015). In some contexts, direct observation is simply not
possible, for example with extinct great apes and human groups. Stable isotope
analysis and dental microwear studies fail to provide total dietary data, and instead
only give a picture of broad dietary patterns such as consumption of particular plant
categories or mechanical properties of diet (Grine, 1986; Scott et al., 2012).
Furthermore, even where direct observational data on food acquisition is available,
data collection is frequently constrained because observation is only feasible over a
short window of the lifetime of an individual that may live up to several decades.

Dental calculus sampled from living or dead individuals is rapidly gaining

recognition as an invaluable material for the reconstruction of life history. Since
Armitage (Armitage, 1975) first recognised plant remains from the teeth of
ungulates, studies have reported starch grains, phytoliths, pollen grains, diatoms,
mineral particles, proteins and DNA from diverse human and animal populations
(Kucera et al., 2011; Adler et al., 2013; Salazar-García et al., 2013; Power et al., 2014b,
2015a; Warinner et al., 2014). Using dental calculus from present day forager-
horticulturalists, Leonard (Leonard et al., 2015) showed for the first time that
recovered microremains also occur in consumed foods verifying the link between
microremains in calculus and diet. As demand grows for dietary history data,
analysis of phytoliths and starches in dental calculus is been increasingly used to

68
reconstruct dietary ecology and ecological niches (Lalueza-Fox et al., 1996; Gobetz
and Bozarth, 2001; Boyadjian et al., 2007; Henry et al., 2012; Mickleburgh and Pagán-
Jiménez, 2012; Salazar-García et al., 2013; Buckley et al., 2014; Lazzati et al., 2015).

Despite the promise of calculus dietary studies, they are hindered by the lack
of research that cross-validates the dietary material recovered in calculus with the
organism’s actual feeding ecology and life history. Until recently, our understanding
of what the plant matter preserved in calculus precisely represents has been
speculative. The initial effort to characterise the microremain record by Leonard
(Leonard et al., 2015) reported that calculus captured only a limited proportion of
dietary breadth. In this study many vegetable foods lacked phytoliths and starches,
and cooking may have significantly reduced starch abundance even if present.
Dietary patterns were established through interviewing and short-term camp stays
by Leonard, and though the recovered microremains corresponded to the average
diet of the population, the dietary records lacked insight into the long term life
history of individuals. Without dietary records that cross intra- and inter-annual
cycles, our knowledge of the nature of the calculus record and its potential for
archaeological studies is incomplete. Furthermore, it is unclear if the calculus dietary
record has input from non-dietary sources (e.g. preparation of plant-based tools and
oral hygiene) as well as the consumption of stomach contents (Buck and Stringer,
2014; Dudgeon and Tromp, 2014; Tromp and Dudgeon, 2015), with bias from
diagenetic and taphonomic factors rendering it ultimately purely stochastic.

In our study, we compare the plant microremains from the calculus of the
chimpanzees of the Taï Forest to 22 years of group averaged dietary observation data
in order to validate the calculus record and explore its potential as a source on
information on life histories. For this purpose, the study of chimpanzees provides
several strengths as a model. First, the chimpanzee mouth is analogous to humans,
in that chimpanzees often accumulate large deposits of calculus unlike some
mammals. Secondly, chimpanzees produce salivary amylase unlike some primates
(Santos et al., 2012) although it is less abundant than in humans (Perry et al., 2007).
Thirdly, Taï chimpanzees have a broad diet that includes nearly all food classes (e.g.
fruits, piths, leaves, mammals, birds, invertebrates and honey) and is thus relevant
to understanding hominin evolution in the African tropics and dietary ecology of
hunter-gatherers living in other tropical regions. Fourthly, chimpanzees only
consume raw foods meaning the microremains in their food are unaltered by
cooking processes.

69
 We sampled calculus from 24 individual chimpanzees using established
methods (Dobney and Brothwell, 1986; Mickleburgh and Pagán-Jiménez, 2012), and
built a random forest model in the R software environment (Breiman, 2001) to
identify the microremains based on multivariate comparison to reference material
(Fenwick et al., 2011; Saul et al., 2012; Out et al., 2014; Coster and Field, 2015; Out
and Madella, 2015) (see detailed methods 7.2). We predict that if microremains
reflect diet, they are accumulative in calculus and should increase with age of the
individual. Chimpanzee sex might also influence the abundance of microremains,
since male and female chimpanzee are known to vary in their time allocation to
different food resources (Wrangham and Smuts, 1980; Boesch and Boesch-
Achermann, 2000; N’guessan et al., 2009; Fahy et al., 2013). We also anticipate that
the proportions of microremains from each plant will be determined by the
frequency with which that plant is consumed and how abundant the microremains
are in the plant tissue. Although we knew the taxonomic identity of the reference
plants at the level of species, an important amount of dietary observation data was
present only at the genus level. Therefore, we performed our analyses at the genus
level in order to have a higher chance of capturing long term dietary averages for the
group, and refer only to genera in the text. Except where otherwise noted, our
analyses were done at the group level observational data, since the records for
individual chimpanzees were not complete enough to provide a detailed overview
of life history. We found the phytoliths in dental calculus to be an approximate
record of diet, and furthermore that microremains can reflect important behaviour
like nut-cracking and episodes of Taï Chimpanzee life history such as the age of
weaning.

4.2 Materials

4.2.1 Taï Forest reference material

A reference collection with 91 genera (113 species) of the most frequently

consumed chimpanzee plant foods in the Taï Forest was collected and examined for
phytoliths and starch (Table 7; Appendix table 4) starches. Phytoliths and starches
were isolated from reference plants using conventional approaches (Piperno, 2006).
We selected thirteen starch- and seven phytolith-producing genera from the 91 we
analysed for the identification model (Appendix table 4).

70
Table 7: Plant genera selected from reference collection species for the predictive identification model with
the microremain content of the dried plant material provided as a percent of dried plant material, and the
frequency of observed consumption provided as number of minutes eaten. We chose to use genus as the
taxonomic rank as some dietary records only identify genus.

Plant category (Genus) Type Plant part % Microremain/Dry Weight Minutes eaten
Elaeis Phytolith Fruit and leaf 4.81 9379
Eremospatha Phytolith Pith 1.72 25,046
Laccosperma Phytolith Pith and seed 2.15 5311
Aframomum Phytolith Seed and leaf 2.13 1704
Sarcophrynium Phytolith Fruit 3.32 1847
Cola Starch Seed 40 35,778
Aframomum Starch Seed 54.58 1704
Piper Starch Seed 39.22 492
Sacoglottis Starch Fruit 2.46 258,225
Panda Starch Seed 0.85 17,299
Coula Starch Seed 31.15 118,095
Napoleona Starch Seed 20.79 51
Gilbertiodendron Starch Seed 23.87 11,808
Eremospatha Starch Pith 2.93 25,046
Calpocalyx Starch Fruit 9.1 49,074
Sarcophrynium Starch Fruit 23.83 1847
Xylia Starch Seed 19.58 46,587
Treculia Starch Seed 23.87 58,093

4.2.2 Calculus sampling

The calculus samples used for our analysis come from permanent and
deciduous molars of 24 chimpanzees from the Taï Chimpanzee Osteology Collection
at the Max Planck Institute for Evolutionary Anthropology (MPI-EVA) with varied
life histories (4.6; Table 8, Appendix table 5). We selected only molars to standardise
the sampling, and chose teeth that were encrusted with a prominent band of
supragingival calculus (calculus present above the original gumline) on the enamel
crown. Deposits of supragingival calculus were present on all individuals ≥ 1 years
old. Subgingival calculus was also present but was not sampled since it occurs below
the former gums and it is unclear if it preserves food remains. Calculus on the teeth
was documented with photography before sampling, and the colour noted with how
each skeleton as treated before our sampling (Appendix table 5). Packing material
was sampled as a control. An unidentified adhesive used in the curation of some
specimens was removed before sampling. A dental scalar was then used to remove
small areas of calculus. The amount of calculus sampled had no relationship with the

71
amount of calculus present on the teeth except in the youngest chimpanzees (<5.3
years), where calculus was rare and almost entirely collected. We sampled in clean
conditions in a laminar flow cabinet at positive-pressure at the MPI-EVA. We then
weighed each of the samples and transferred them to microcentrifuge tubes. After
sampling, the teeth and surviving calculus were photographed.

Some studies have highlighted the risks of laboratory contamination from

modern plant microremains (Crowther et al., 2014; Barton and Torrence, 2015;
Weyrich et al., 2015). To address the possibility of contamination, we conducted a
regime of weekly laboratory cleaning to remove contamination. All work surfaces
were wiped with hot water, washed with starch-free soap and wiped with 5 %
sodium hydroxide (NaOH). We additionally performed wipe tests before and after
weekly cleaning to quantify starch contamination and assess contaminating types.
Wipe tests retrieved settled particles of the surface area (74 x 43 cm 2) of the
laboratory positive-pressure laminar flow hood used for mounting. Results of these
intensive contamination control tests are in Power and colleagues (2015b).

72
Table 8: All chimpanzee dental calculus samples analysed.

Name ID Tooth Sex Weight (mg) Age at death

In years In months
Ophelia 14993 Lower Left DM1 F 0.025 1 12
Leonardo 13432 Upper DM2 M 0.329 1.92 23
Bambou 11777 Lower Left DM1 M 0.135 2.08 25
Piment 11788 Lower Right DM1 F 0.27 3.58 43
Oreste 14995 Lower Left M1 M 0.536 5.25 63
Hector 12175 Upper Right M1 M 0.689 5.67 68
Noah 15011 Lower M1 M 1.165 7 84
Lefkas 13433 Upper Left M2 M 0.595 7.58 91
Tina 11790 Lower Right M1 F 1.36 9.08 109
Dorry 15020 Lower Right M2 F 0.742 11 132
Zerlina 11792 Lower Left M3 F 0.878 12.3 144
Clyde 11779 Lower Right M1 M 1.131 13 156
Agathe 11775 Lower Right M2 F 6.076 16 192
Leo 15012 Lower Right M3 M 1.085 19 228
Bijou 11778 Lower Left M2 F 5.041 19 228
Castor 13439 Lower Left M1 F 6.982 22 264
Kendo 11781 Lower Left M2 M 2.895 25 300
Fanny 11780 Lower Left M3 F 3.915 25 300
Venus 15001 Upper Right M1 F 1.133 27 324
Goma 15004 Upper Right M3 M 13.208 28 336
Rubra 15023 Lower Left M2 F 6.751 38 456
Ondine 11786 Lower Left M1 F 1.529 39 468
Mkubwa 13435 Lower Right M3 M 0.324 40 480
Brutus 15029 Upper Left M3 M 3.246 46 552

4.3 Methods

4.3.1 Optical microscopy

Optical microscopy was performed at the Plant Foods in Hominin Dietary

Ecology laboratory in the MPI-EVA (for reference collection microscopy see 4.6 and
Appendix table 6). We added 150 μl of 10 % hydrochloric acid to the calculus sample
for one to three hours. The samples were then centrifuged at 1691 x g (Heraeus
MEGAFUGE 16 with a microcentrifuge rotor) for 10 minutes and then about 100 μl
of supernatant was decanted and replaced with distilled water. This was repeated
three times to remove the hydrochloric acid. After the second decanting, it was
refilled with a 25 % glycerine solution. The sample was then ground in the solution
in the 1.5 ml Eppendorf microcentrifuge tube to reduce sample loss due to static

73
electricity. The samples then were centrifuged again at the same speed, and about 1
ml of supernatant was decanted. We mounted 20 μl per slide on as many slides as
needed in order to examine the entire sample. Microscopy was used as in
conventional phytolith and starch studies (Power et al., 2014a; b). We examined each
slide under bright field and cross-polarized light on a Zeiss Axioscope microscope at
400 × magnification. We photographed each microremain and described each with
the international microremain nomenclature including the International Code of
Phytolith Nomenclature (Madella et al., 2005). In some cases, starch aggregates were
identified in calculus. In this case, each component grain of each aggregate was
counted as an individual starch.

4.3.2 Microremain identification

We identified microremains with a reference collection using multivariate analysis

with a random forest algorithm. We collected five general microremain
measurements and four specific to phytoliths and six specific to starches from a total
of 900 reference microremains (Table 7). With the reference collection generated
(Appendix table 6) we generated a certainty score that each matched each reference
collection genus. The validity was tested through cross-validation with a subset of
reference data. We identified the microremain as coming from the genus with the
highest certainty score.

4.3.3 Behavioural records

The chimpanzees of the Taï Forest have been studied since the
commencement of the Taï Chimpanzee Project in 1979 (Boesch and Boesch-
Achermann, 2000). The detailed recorded behaviour of the group included
observation of feeding time and food item consumed. The feeding records used in
our study span the period between 1992 and 2014. The database includes 1,165,150
million behavioural observations of about 128 chimpanzees, with a total of 417,628
dietary observations (2,380,202 minutes). However, only roughly 30,000 observations
come from chimpanzees available for sampling at the osteology collection.
Furthermore, most of these chimpanzees have only sporadic coverage of their life
history. Therefore, instead of using dietary records of individual chimpanzees or the

74
collated records of the 24 chimpanzees we sampled, we chose to combine dietary
records from all 128 chimpanzees to best represent the average Taï Forest diet.

The feeding record includes the times when a chimpanzee started and
finished eating, and the food consumed. We chose only those feeding records where
the genus of the plant food eaten was documented, and calculated the total amount
of time spent consuming each resource. Behavioural records do not account for
variations in the volume of food consumed in given number of minutes. In addition,
although some observations record the specific plant part that was eaten, most do
not, so we do not include this information.

4.3.4 Statistical analysis

To test for the effects of age on microremains we used a negative binomial

regression (log link) with a count of each microremain class treated as a response
(phytoliths, starches and other unsilicified plant fragments) using a likelihood ratio
test in R 3.1.0 (R Core Team, 2014). We ran the regression using the glm.nb function
of R package MASS (Venables and Ripley, 2002). The full model included the fixed
effects of age and sex (4.6). The mg weight of each calculus sample was used to
weigh the model to account for larger samples likely being more representative of
overall diet due to the potential of microremains to have a clustered distribution in
the calculus matrix. Controlling for weight, heavier samples have less variable
microremain counts (Compare Table 8 with Fig. 10). The full model was compared
with a null model using an ANOVA. We used likelihood ratio tests comparing the
full models with reduced models in which each fixed effect was dropped, one at a
time. Model assumptions were met. Collinearity was not an issue (largest Variance
Inflation Factor=1.001) and leverage values as well as DFBeta values indicated no
obvious highly influential cases.

To explore the relationship between diet and the phytolith microremains

found in dental calculus we tested an observational random effect Poisson model
with likelihood ratio tests. We used counts of each genus predicted to be present
with the total minutes spent consuming each genus. For this, we used the glmer
function of the R package lme4 (Bates et al., 2013). If any genus was not predicted to
be present in a chimpanzee sample, they were included as a 0 value. Our full model
included minutes and chimpanzee age in months as fixed effects, and sex as a
control predictor. The model included the weight of each calculus sample and the

75
successful identification rate of each type of genera as model weights, and used
microremain content as an offset to factor in significant differences in content
between different genera. To prepare the data, we z-transformed the minutes and
age variables. The chimpanzee individual was included as a random slope term
while year of death, tooth and food type were treated as random intercept terms. An
id was assigned to each observation, and this was also included as a random
intercept, thus reducing overdispersion to (x2=13.369, df=116, dispersion
parameter=0.115) in the phytolith model. To test the significance of the full model it
was compared with a null model excluding fixed effects of minutes of observation
and age. Variance inflation factors (VIF) were derived to assess collinearity using the
function vif of the R-package car, from a standard linear model minus random
effects, as offsets and weights (Fox and Weisberg, 2002; Field, 2005). Variance
inflation factors indicated collinearity to not be an issue (largest VIF=1.02). We tested
model stability by excluding each random effect one by one from the data set,
running the full model and comparing the results with those from the original model
that suggest no highly influential cases.

To explore the relationship between diet and the starch microremains we

could not use the same approach due to high zero inflation present in the starch
data. To overcome this, we implemented a mixed effects logistic regression using the
same terms, random effects, weighs and offset as the phytolith Poisson model. This
required the counts data (the response) to be treated as presence and absence data
resulting in some loss of data. Variance inflation factors (Field, 2005) were derived to
access collinearity using the function vif of the R-package car, from a standard linear
model minus random effects as well as offsets and weights (Fox and Weisberg,
2002). Variance inflation factors indicated collinearity to not be an issue (largest
VIF=1.018).

4.4 Results

4.4.1 Identification of the microremain assemblages

We were able to examine 91 of the 157 genera (113 of 230 species) of plants
that the 128 Taï chimpanzees consumed during the observation period. After
assessing these plants, we noted thirteen starch-producing genera that could be
included in our identification model. Unlike starches, phytolith were abundant in
most plants in many different morphotypes. Not all morphotypes could be included

76
in analysis, so we choose one morphotype (globular and spheroid) and identified the
five genera that produces them (Table 7; Fig. 9). Most of these thirteen starch-
producing and five-phytolith producing genera are major sources of food for these
chimpanzees (Appendix fig. 2).

For each microremain-producing plant genus, we collected data from 50

microremains, to provide a range of measurements within each genus. We collected
nine types of measurements for phytoliths and 11 for starches from 900
microremains using Zeiss AxioVision Microscopy Software (Appendix table 6). By
using a subset of the reference collection to test the model, we assessed the success
rate of identification of each genus with the model (Appendix table 8, Appendix
table 9). Some genera were reliably identified, and others were more difficult to
identify. For example, Sarcophrynium phytoliths were identified successfully 94 % of
the time while Panda starch was only identified 22 % of the time. Generally,
phytoliths were identified more reliably (Appendix table 8, Appendix table 9). Using
this random forest model, we were able to proceed with identification of
microremains recovered from the calculus.

77
Fig. 9: Starch and phytolith morphotypes used in the identification model. Each scale bar represents 10 µm. (a)
Aframomum sceptrum seed phytolith, (b) Aframomum excarpum leaf (b) Aframomum excarpum leaf phytolith,
(c) Aframomum excarpum seed starch under bright field (left) and cross polarized light (right), (d)
Laccosperma opacum pith phytolith, (e) Laccosperma secondiflorum seed phytolith, (f) Calpocalyx sp. fruit
starch under bright field (left) and cross polarized light (right), (g) Cola nitida seed starch under bright field
(left) and cross polarized light (right), (h) Coula edulis seed starch under bright field (left) and cross polarized
light (right), (i) Elaeis guineensis fruit phytolith, (j) Elaeis guineensis leaf phytolith, (k) Gilbertiodendron
splendidum seed starch under bright field (left) and Cross polarized light (right), (l) Eremospatha macrocarpa
pith phytolith, (m) Eremospatha macrocarpa pith starch under bright field (upper right) and cross polarized
light (lower left), (n) Napoleona leonensis seed starch under bright field (left) and cross polarized light (right),
(o) Panda olesosa seed starch, (p) Piper guineense seed starch under bright field (upper right) and cross
polarized light (lower left), (q) Sacoglottis gabonensis fruit starch under bright field (left) and cross polarized
light (right), (r) Sarcophrynium prionogonium fruit phytolith, (s) Sarcophrynium prionogonium fruit starch
under bright field (left) and cross polarized light (right), (t) Treculia africana seed starch under bright field
(left) and cross polarized light (right), (u) Xylia evansii seed starch.

78
 Of the 24 chimpanzee calculus samples, we found starches in 17 of the
samples, and phytoliths in 20 (Fig. 10, Fig. 11; Table 8, Appendix table 10). We also
found unidentified phytoliths, unsilicified plant fragments, diatoms, pollen and
insects, but these were not identified to taxon (Table 8). In ambiguous cases
microremains were classified as possible starches and specifically stated, but were
not used for statistical genera identification. Most definite starches and phytoliths
that were free from damage (234 starches and 1035 phytoliths) were identified to
genus using the random forest model, which assigned each unknown microremain
to a genus and provided a certainty score that indicated the confidence with which
that assignment was made. A microremain was considered to be damaged if it
showed pitting, ruptured surfaces or other major irregularities. The highest certainty
score for each individual microremain depended heavily on each genus
identification rate (as described above), but generally ranged between 0.25 and 0.95.

79
 Unsilicified plant remains Starches Phytoliths Age
a
Breastfeeding Post-weaning
300 50
45
250
40
Microremains numbers

35
200

Chimpanzee age (years)

30
150 25
20
100
15
10
50
5
0 0
b 50
250
45

40
200
35
Microremains per mg

30

Chimpanzee age (years)

150
25

20
100
15

50 10

0 0
Hector

Leo
Tina

Zerlina

Castor
Bambou

Lefkas

Fanny

Ondine
Piment

Rubra

Brutus
Noah
Oreste

Clyde
Dorry

Agathe
Ophelia
Leonardo

Goma
Venus

Mkubwa
Bijou

Kendo

♀ ♂ ♂ ♀ ♂ ♂ ♂ ♂ ♀ ♀ ♀ ♂ ♀ ♂ ♀ ♀ ♂ ♀ ♀ ♂ ♀ ♀ ♂ ♂

Fig. 10: Microremains recovered in calculus samples. Microremains include Unsilicified microremains,
starches (definite and possible) and phytoliths recovered with chimpanzee age at death (in years) and
approximate age of the cessation of weaning highlighted. a=total counts and b=counts per milligram of
calculus. The number of microremains per mg in Ophelia was affected by an unusually small amount of
calculus in the sample.

80
100%

80%
Sarcophrynium
60% Aframomun

40% Laccosperma
Elaeis
20% Eremospatha
0%

Leo
Tina
Hector
Noah
Lefkas

Castor
Fanny

Rubra
Bambou

Zerlina

Ondine

Brutus
Clyde
Piment

Goma
Dorry
Ophelia
Leonardo

Oreste

Agathe
Bijou

Kendo
Venus

Mkubwa
100%
Xylia
80% Sarcophrynium
Gilbertiodendron
60% Calpocalyx
Coula
40% Treculia
Eremospatha
20% Napoleona
Panda
0% Sacoglottis
Piper
Leo
Hector

Lefkas
Tina

Zerlina

Ondine
Castor
Fanny
Bambou

Agathe
Piment

Noah

Dorry

Brutus
Clyde

Rubra
Goma
Oreste
Ophelia
Leonardo

Kendo
Venus

Mkubwa
Bijou

Aframomun
Cola

Fig. 11: Microremain assemblages recovered in calculus. (top) Bar chart of the composition of the phytolith
assemblage recovered from calculus. (bottom) Bar chart of the composition of the starch assemblage recovered
from calculus. The individuals are ordered by age from youngest to oldest.

4.4.2 Assessment of biases in our data

First of all, it was important to ascertain if the treatment of the skeletal

material to prevent the spread of disease (including one year of burial, and various
chemical treatments) had impacted microremain preservation in the calculus. After
2004 chorine and formalin was used to clean skeletal material. Boiling may have
been used on some skeletons to clean them and remove Ebola pathogens between
the Autumn of 1994 and the Spring of 1996 (Appendix table 10). To test if the three
types of treatments significantly influenced starch preservation we used a Kruskal-
Wallis test starch per mg on samples from each period (H=3.7633, df=2, p-
value=0.1523). We included microremains classified as possible starches in the starch
per mg count (Appendix fig. 1). Due to the small sample size we calculated a
Kruskal-Wallis p-value based on 999 random permutations. This indicated no
differences between the three groups (Permutation H=7.1215, df=2, p=0.159).

81
Previous studies of other organic material (bone collagen) in the Taï skeletal
collection have indicated no significant post-mortem alteration (Fahy et al., 2013,
2014). While collagen does not necessarily behave in the same manner as plant
microremains, it is likely that the comparable hydroxyapatite mineral matrices of
bone and calculus have a similar protective effect on the organic materials trapped
within them (Nicholson, 1996).

Before comparing the calculus results to the observational records, we wanted

to see if there was excessive variation in plant representation among the calculus
samples. Phytoliths from four of the five phytolith-producing genera were found on
most, but not all, of the calculus samples, suggesting that there is not much
variability among these calculus samples (Fig. 11, Appendix table 11). Some genera
are found in each sample (Eremospatha and Elaeis) while others, like Sarcophrynium,
were rare (Appendix table 12). However, the starch record varies significantly
among individuals, with most of the thirteen starch-producing genera seldom found.
This probably reflects the far lower numbers of starches compared with phytoliths.
Several genera dominate the starch record, namely Gilbertiodendron, Coula,
Eremospatha, Treculia and Cola (Fig. 11, Fig. 12). Most microremains were isolated, but
three calculus samples had four starch aggregates from Piper; each starch in the
aggregate was counted as an individual starch grain and thus constitutes a large
proportion of the total number of the recovered starches. This potentially biases the
starch assemblage’s dietary representativeness (Fig. 11; Appendix table 12). In sum,
it seems that there is not much variation in the phytolith record of our chimpanzee
samples, but the starch record is less homogeneous. Another potential source of bias
comes from the differential preservation of microremains relating to their inherent
properties, like size and shape. We noted that our results were biased towards foods
with larger sized microremains. Elaeis phytoliths and Cola starches, the largest
microremains in the study (Fig. 11, Fig. 12), are disproportionately frequent across
the assemblages even after controlling for the high concentration of microremains
within these genera. They are frequently found, but are not dominant foods
(Appendix fig. 2).

82
 a

Laccosperma
Aframomum
Elaeis
Eremospatha
Sarcophrynium

b Cola
Aframomum
Piper
Sacoglottis
Panda
Coula
Napoleona
Gilbertiodendron
Eremospatha
Calpocalyx
Sarcophrynium
Xylia
Treculia

Fig. 12: Plant genera represented by microremain assemblages and Chimpanzees diet. Microremain counts are
normalised by dividing counts by the percent content of by dry plant weight of starches and phytoliths
among different genera. (a) Phytolith counts compared with feeding records. Outermost ring=proportions of
minutes spent consuming each genus averaged across the feeding records of sampled 24 chimpanzees,
middle=proportions of minutes spent consuming each genus averaged across the feeding records of all 128
chimpanzees, innermost=phytolith counts from the sampled 24 chimpanzees. (b) Starch counts compared
with feeding records outermost ring=proportions of minutes spent consuming each genus averaged across the
feeding records of sampled 24 chimpanzees, middle=proportions of minutes spent consuming each genus
averaged across the feeding records of all 128 chimpanzees, innermost=starch counts.

83
 4.4.3 Microremain accumulation, chimpanzee age and sex

If microremains reflect diet they should be more abundant in older

chimpanzees, and might vary by sex. We tested this using (see detailed methods
below). We ran separate tests for phytoliths, starches and unsilicified remains. For
phytoliths, the full model of age and sex significantly influenced the count of
phytoliths (x2=11.794, df=2, P=0.0003), and the effect of age was also significant by
itself (x2=12.753, df=1, P=0.0004) (Appendix table 13). Older chimpanzees generally
have a higher abundance of phytoliths. However, sex by itself did not explain the
abundance of phytoliths we found (x2=0.028, df=1, P=0.866). For unsilicified
microremains, age and sex as the full model significantly influenced the
microremain count (x2=10.067, df=1, P=0.015), the effect of age alone was also
significant (x2=9.202, df=1, P=0.0015), but not sex by itself (x2=0.59, df=1, P=0.806).
Starch abundance was significantly determined by age and sex together (x 2=23.994,
df=2, p=6.1622e-06). Older chimpanzees generally have a higher abundance of
starches (x2=3.559, df=1, p=0.0592). Unlike with phytoliths and unsilicified remains,
sex strongly influenced the abundance of starch, with female chimpanzees having
more starches (x2=17.301, df=1, p=3.1897e-05) (Appendix table 13).

4.4.4 Microremain dietary picture and observational feeding records

We predicted that more frequently consumed plants should be highly

represented in the chimpanzee calculus. To test this, we used an observational
random effect Poisson model (4.3.4). The count of identified classes of microremains
(phytoliths and starches) belonging to a particular genus was our response variable,
and the fixed predictors were: (a) minutes spent consuming each genus, and (b)
chimpanzee age in months. Sex was included as a control predictor, and both
calculus sample weight and successful identification rate of each genus were
included as weights. We used counts of each genus predicted to be present with the
total minutes spent consuming each genus. The chimpanzee individual was
included as a random slope term, while year of death, tooth and food type were
treated as random intercept terms (see methods below for 7.2).

When comparing the genera proportions present in diet (calculated as the

number of minutes spent foraging on a genus) with the recovered phytolith
assemblages, we found a clear relationship. The number of minutes spent
consuming a given plant genus influences its phytolith count in the calculus

84
assemblage even when accounting for the effects of sex, the tooth we sampled,
variation in phytolith production between different plants, and the year the
individual died. More specifically, an increased reliance on a genus leads to an
increase in its representation in calculus (x2=4.048, df=1, P=0.045; Appendix table 13;
Fig. 13). The age of the chimpanzees was found to not influence how well it matches
group diet (x2=0.356, df=1, P=0.55; Appendix table 13).

Fig. 13: Mixed Poisson regression model predicted values. The number of phytoliths from a genus increased
as the minutes spent consuming this plant resource increased.

In contrast to phytoliths, there was no significant effect of consumption time

on starch numbers (x2=1.95, df=2, P=0.376). The number of minutes this group spent
eating a specific genus of starchy foods does not influence its frequency in dental
calculus. Yet there is an element of uncertainty because starches vary more among
individuals than do phytoliths, as described above, and do not seem to be as good a
record of dietary behaviour. Fig. 10 and Fig. 11 show the discrepancy between the

85
consistency of starch and phytolith results clearly. These results may be a product of
post-mortem diagenesis that influenced our chimpanzee samples, including burial to
deflesh the remains (Appendix table 5).

4.4.5 Weaning and other behavioural signatures in calculus

The microremains in the Taï calculus record other aspects of their behaviour.
First, microremains were strikingly rare in samples from individuals less than 5.3
years old (Fig. 10, Fig. 11; Appendix table 5). The calculus deposits were sparse on
these individuals, but despite the small volume of calculus, it was notable that only a
single starch and an unsilicified plant fragment were found in these samples.
Chimpanzees more than 5.3 years old typically show high numbers of microremains,
regardless of the size of the calculus deposit.

Second, the exact plants that were recovered in the calculus provide an
interesting view of an important learned behaviour. In our sample, many calculus
samples had starches from the Coula nut, which is mainly consumed once
chimpanzees learn to crack open these nuts. Coula nut starches were found in
samples from individuals across all age ranges (except those under 5.7 years) (Fig.
14). Although common, Coula nut appears to be underrepresented in our sample. It
was found only in nine calculus samples, despite this plant being a major food
source, comprising of 4.7 % of the total Taï diet.

86
 Coula Chimpanzee age (In years)

7 50

45
6
40

Chimpanzee age (in years)

5 35
Coula starch count

30
4
25
3
20

2 15

10
1
5

0 0
Ophelia

Piment

Agathe
Oreste

Lefkas

Clyde

Leo

Goma

Ondine

Brutus
Bambou

Tina

Castor

Venus

Mkubwa
Hector

Dorry

Kendo

Rubra
Fanny
Bijou
Noah
Leonardo

Zerlina

Fig. 14: Abundance of Coula nut starches with chimpanzee age at death. Coula nut consumption requires nut
cracking and its presence implies nut cracking and tool use or food sharing. The individuals are ordered by
age from youngest to oldest.

4.5 Discussion

Much of the chimpanzee calculus is relatively rich in plant microremains

compared with what has been reported in previous studies of human calculus
(Kucera et al., 2011; Mickleburgh and Pagán-Jiménez, 2012; Leonard et al., 2015).
This is not entirely unexpected for several reasons. First, our samples are modern,
and post-mortem microremain diagenesis is therefore less acute than in ancient
remains. Secondly, Taï Chimpanzee diet is uncooked, plant-dominated and
voluminous (Appendix fig. 1; Appendix fig. 2). Thirdly, and in contrast to humans,
chimpanzees consume a large amount of phytolith-rich material. This richness in
microremains is largely confined to phytoliths. Starch abundance falls within ranges
observed elsewhere (Mickleburgh and Pagán-Jiménez, 2012; Power et al., 2014b;
Leonard et al., 2015).

87
 It is evident that starches are underrepresented and in some samples are even
totally absent. In addition, phytoliths present a far more uniform picture of diet
between different chimpanzees. This may be due to diagenesis occurring during the
preparation of the skeletons for the osteology collection. It is known that all
skeletons were buried for short periods of time during the defleshing process
(Appendix table 5). These processes may preferentially alter or remove starches from
the calculus record that are not sufficiently mineralised and sealed, while leaving the
phytoliths relatively unaltered. Yet, our Kruskal-Wallis test indicates cleaning
processes have not influenced starch numbers.

The comparison of microremains with diet was possible because our

methodology generated a deep profile of dietary history. However, our metric of
food use (minutes spent consuming a food) is not identical to the total volume of
consumed food by each chimpanzee, or a food’s total energetic value. Unfortunately,
our analysis was not able to account for this.

Additionally, we found that the microremain record was likely biased by the
differential survivability of microremain from different plants. The plants with the
largest starches and phytoliths were overrepresented in our sample, possibly due to
the larger surface area. This ties in with research that shows that phytolith and starch
morphology and surface area is linked to long term stability (Haslam, 2004; Cabanes
et al., 2011). Larger blocky microremains may be preferentially preserved. This is
noteworthy given that their larger surface area would enhance their contact with
bacterial and chemical process and alteration.

Overall, our results verify that the calculus record can be accumulative by
showing that older individuals present more microremains. Sex may be a factor to
take into consideration, and seems to influence the accumulation of starches but not
phytoliths or unsilicified remains. It may reflect higher consumption of starches by
female chimpanzees, or sex differences in amylase production or calculus formation,
as has been suggested for humans (Monteiro da Silva et al., 1998). We do not
currently have the ability to distinguish among these possibilities. The increase in
microremains with age and possibly sex implies that microremain accumulation is
bound up in aspects of diet that regulate calculus formation. Thus, microremain
presence and proportions are likely effected or confounded by all the factors that
influence plaque and calculus such as the intake of protein, smoking, polysilicic acid
and silica (Damen and Ten Cate, 1989; Roberts-Harry and Clerehugh, 2000; Jin and

88
Yip, 2002). Calculus clearly can approach a long-term dietary signal, although the
timespan involved is not yet clarified.

Our results strongly suggest that care must be taken when interpreting the
microremains record preserved in dental calculus, particularly the starch grain
record. However, our results also indicate that microremains in calculus can be used
to recover important information about diet, behaviour and life history. For example,
we observed a lack of microremains from individuals sampled from deciduous teeth
of chimpanzees less than 5.3 years old. The microremain assemblages could indicate
a rapid accumulation of microremains as solid food enters the diet (Fig. 10). This
pattern matches what is generally reported for age of weaning using other measures.
Much information on the age of chimpanzee weaning is estimated from inter-birth
interval (Fahy et al., 2014). Inter-Birth Interval estimates of weaning ages vary from
4.5 years at Gombe to 5.75 years at Taï (Boesch, 1997), to 6 years at Mahale (Nishida
and Hasegawa, 1992). Yet inter-Birth Interval is an indirect measure as it includes
more than simply suckling duration. Isotopic based data indicates weaning at Taï
commences at 2 years and ends at 3-4.5 years varying by factors such as sex of the
offspring. If we combine this infant microremain signal together with the verification
of the accumulative nature of the microremain assemblages, we can conclude
calculus reflects information on the weaning transition that may be useful for
studying unhabituated populations. Researchers should expand this research to
infants from recent foragers and horticulturists to develop its applicability.

Furthermore, though the starch dietary record appears more stochastic than
that of phytoliths, starches can still provide useful information about behaviour.
Many of our starches come from the Coula nut (Fig. 11, Fig. 14). Among
chimpanzees, Coula consumption requires a learned behaviour: nut cracking with a
hammer and anvil. This behaviour is restricted to a limited area of the chimpanzee
range in West Africa (Boesch et al., 1994). The presence of Coula starches (Fig. 11, Fig.
14) shows calculus can reveal nut-cracking behaviour in a group. The fact that tool-
use in a group is discernible is relevant for dental calculus studies in both
primatology and hominin evolution. The use of Coula nut is influenced by age and
sex differences in nut cracking (Boesch and Boesch, 1984; Boesch and Boesch-
Achermann, 2000), and, as expected, Coula starches are absent in youngest
chimpanzees who are not yet weaned. Even after weaning infant nut consumption is
low and is derived from nuts cracked by the mother as it takes years to learn how to
crack nuts (Boesch and Boesch-Achermann, 2000). Beyond this, we do not have

89
enough calculus samples to examine if there are sex or age differences in the calculus
record of nut cracking.

This profile of Taï Chimpanzee diet reflects high amount of contextual

information available on a single population. Much research utilizing dental calculus
is interested in reconstructing diet of collections of individuals and not
contemporaneous populations from archaeological sites. Researchers using
archaeological populations with small and even large samples must be aware of the
unlikeness of being able to capture the full dietary breadth with dental calculus. The
best strategy to account for this issue is to maximize the number of samples in a
study.

In summary, the study verifies the relevance of dental calculus for

investigations on diet, food acquisition behaviour and life history. It is the first to
link dental calculus with foods that entered the oral cavity in quantified abundances
but it also identifies clear weaknesses of this method. The data also provide valuable
information on the commencement of plant food consumption in wild chimpanzees,
and confirms the consumption of solid foodstuffs from at least 5.3 years in life. Our
study suggests that calculus analysis provides a rich but wavering insight into
complex dietary structure, and that phytoliths, when present in calculus and in diet,
may provide a more reliable record of diet than starches.

90
 5

Dental calculus indicates widespread plant use within

the Neanderthal dietary niche
Robert C. Power,1 Domingo C. Salazar-García,2, 3, 4 Mauro Rubini,5, 6 Katerina Harvati,7
Andrea Darlas,8 Michael Walker,9 Jean-Jacques Hublin,4 Amanda G. Henry1

1Max Planck Research Group on Plant Foods in Hominin Dietary Ecology, Max Planck Institute for
Evolutionary Anthropology, Deutscher Platz 6, 04103 Leipzig, Germany.
2Department of Archaeology, University of Cape Town, Cape Town, South Africa.
3Departament de Prehistòria i Arqueologia, Universitat de València, València, Spain.
4Department of Human Evolution, Max Planck Institute for Evolutionary Anthropology, Leipzig,
Germany.
5Department of Archaeology, University of Foggia, Italy.
6Anthropological Service of S.B.A.L. (Ministry of Culture Italy), v. Pompeo Magno 2, Rome, Italy.
7Paleoanthropology, Department of Early Prehistory and Quaternary Ecology, Senckenberg Center
for Human Evolution and Paleoecology, Eberhard Karls University of Tübingen, Rümelinstrasse 23,
Tübingen 72070, Germany.
8Ephoreia of Paleoanthropology and Speleology of Northern Greece, Greek Ministry of Culture,
Navarinou 28, 55133 Kalamaria, Thessaloniki, Greece.
9Departamento de Zoología y Antropología Física, Universidad de Murcia, Murcia, Spain.

Submitted to the Journal of Human Evolution.

Abstract

The ecology of Neanderthals is a pressing question in the study of hominin

evolution. Diet appears to have played a prominent role in their adaptation to
Eurasia. Isotope and zooarchaeological studies indicate that Neanderthals ate large
quantities of meat, and that there was little variation in their diet across Eurasia.
However, we have only a fragmentary picture of their dietary ecology and how it
may have varied among habitats, because we lack detailed information about their
use of plants and other foods. To address the problem, we examined the plant
microremains in Neanderthal dental calculus from five archaeological sites from the
northern Balkans, and the western, central and eastern Mediterranean. The
recovered microremains revealed the consumption of a variety of non-animal foods,
including starchy plants. Using a modelling approach, we explored the relationships

91
among the diversity of microremains with chronological, climatological and
ecological variation. We find no evidence that plant use is confined to the southern-
most areas of Neanderthal geographic distribution. Although Neanderthals were
predominately big game hunters, evidence of diet from dental calculus indicates that
plant exploitation was a widespread and deeply rooted subsistence strategy. Given
the limited dietary variation across Neanderthal range in time and space in both
plant and animal food exploitation, we argue that vegetal consumption was part of a
generally static dietary niche.

5.1 Introduction

Neanderthals occupied a variety of environments drastically different from

those where hominins first evolved, earning them a special position in evolutionary
history. The ability of this hominin to settle in environments as diverse as the
Mediterranean margin and steppe as cold as present-day Arctic tundra implies that
Neanderthals were successful at adapting to new conditions. In particular, their diets
must have been flexible enough to allow them to thrive in these varied
environments. However, some researchers have linked the displacement of
Neanderthals at the end of the Middle Palaeolithic to narrower diets than those from
Upper Palaeolithic peoples (Hockett and Haws, 2003, 2009; O’Connell, 2006). In this
view, Neanderthal subsistence was reliant on a more restricted range of staples than
that of modern humans, giving them a competitive disadvantage against Upper
Palaeolithic peoples.

Dietary breadth models, borrowed from the framework of behavioural

ecology, have provided a means to dissect Palaeolithic dietary adaptations. These
models are predicated on the idea that foragers will select the foods that provide the
most nutritional benefit at the lowest costs, within the constraints imposed by the
environment. The costs and benefits of food are predominantly measured in calories
(Winterhalder and Smith, 2000), or other currencies such as macro or micronutrients
(Rothman et al., 2006). When the return rates for preferred foods decrease, due to
climate change or hunting pressure caused by a population increase, then more food
types are added to the diet. A broadening diet is therefore not an adoption of an
improved diet; just a response to scarcity of preferred food types.

Neanderthals are often interpreted as narrow spectrum foragers (Kuhn and

Stiner, 2006; O’Connell, 2006; Stiner and Kuhn, 2009; Stiner, 2013). Models of Middle

92
Palaeolithic dietary ecology suggest that they hunted predominantly medium and
large prime-age fauna with only infrequent use of small mammals, and aquatic
resources and plant foods (Hockett and Haws, 2005). Nitrogen stable isotope ratios
indicate that they were at the top of the terrestrial food web and obtained most of
their total dietary protein from animal sources, (Richards et al., 2000; Lee-Thorp and
Sponheimer, 2006; Richards and Trinkaus, 2009; Salazar-García et al., 2013; Wißing et
al., 2015). Some zooarchaeologists argue that this diet was stable over time, with
little evidence of a chronological trend towards more diverse resource use (Stiner et
al., 2000; Stiner, 2013). Surviving tool repertoires show scant evidence for the
investment in specialised technology for collecting plants, fish, birds and small
mammals (Kuhn and Stiner, 2006; O’Connell, 2006; Henry et al., 2014), indicating an
unchanging and narrow dietary niche. A low diversification in food choice and high
consumption of large and medium-sized game matches evidence from site density
and their genetic history that imply sparse and, dispersed populations of
Neanderthals that did not deplete high-ranked prey items (Stiner, 1999; Stiner and
Munro, 2002; Macdonald et al., 2009; Verpoorte, 2009; Castellano et al., 2014).

This view of rigid Neanderthal diets is complicated by recent studies

suggesting evidence for variation in diets. Prey selected by Neanderthals varies
throughout their range, often along ecological gradients. In southern regions, there is
evidence for the consumption of hard-to-catch, low-ranked small game (Stiner, 1994;
Blasco and Fernández Peris, 2009; Stiner and Kuhn, 2009; Hardy et al., 2013; Salazar-
García et al., 2013; Fiorenza, 2015). In southern Iberia and western Italy, there is also
zooarchaeological evidence of a contribution of marine resources (Stiner, 1994;
Stringer et al., 2008; Zilhão et al., 2010). A preponderance of low–ranked small game
including shellfish and tortoise (Testudo spp.) is also known from sites in Greece,
Italy, Spain and Israel (Stiner, 1994; Cortés-Sánchez et al., 2011; Blasco and
Fernández Peris, 2012; Sanchis, 2012; Harvati et al., 2013). A study of tortoise
remains at Nahal Meged showed a decrease in size due to hunting pressure and
climate, beginning in the late Middle Palaeolithic, suggesting that Neanderthals
were collecting these foods at significant enough rates to reduce their body size
(Stiner et al., 2000). In Cova del Bolomor, tortoises, rabbits and birds appear to have
been foraged during MIS 6 (Blasco and Fernández Peris, 2009; Salazar-García et al.,
2013). In the warm MIS 5e interglacial, a greater proportion of small game is
observed at several northern European sites despite the apparent continued
dependence on large game (Gaudzinski-Windheuser and Roebroeks, 2011).

93
 The current debate between a rigid, narrow diet and a more variable range of
diets continues because most of our dietary evidence is fragmentary. Large parts of
diet are poorly known, especially plant foods. Recent foragers in northern
environments provide a poor reference for Pleistocene foragers, in part because the
treeless biomes of the Pleistocene have no analogue in the modern era (Stewart,
2005). The biomass of Pleistocene grasslands far exceeded the biomass of present day
Eurasian tundra, providing a greater number of available animals for Neanderthals.
We know less about the productivity of plant foods in this ecological zone
(Verpoorte, 2009), but energy-rich plants were available on the steppe-tundra and
throughout western Eurasia (Sandgathe and Hayden, 2003; Hardy, 2010; Pryor et al.,
2013; Power et al., 2016).

Relatively little evidence of plant use in this context is available. Most isotopic
profiles conducted so far have been produced from collagen, and thus reveal little
information on the consumed macronutrients other than proteins that could have
been obtained from vegetable resources. Macrobotanical remains that survive in a
number of archaeological sites alleviate the gap in the scholarship, but surviving
traces of plant use have limited interpretative power due to taphonomic bias (Weiss
et al., 2004). The most comprehensive studies of dietary variability that incorporate
plant foods stem from indirect lines of evidence, in particular dental wear studies.
Macro- and microwear studies of dental surfaces have revealed that Neanderthals
predominantly consumed meat, with a possible increased use of plant foods in the
southern wooded parts of their range (El Zaatari et al., 2011; Fiorenza et al., 2011).
Microwear of Neanderthals who inhabited cold-steppe environments resembled that
of recent historic Fuegians who inhabited Patagonian cold wet scrublands (Grine,
1986; Fiorenza et al., 2011). However, dental wear is silent on the number and types
of plants consumed, or if low- ranked foods were consumed, meaning these studies
create an incomplete picture of dietary ecology in different environments.

Neanderthals appear to have broader diets in southern regions possibly due

to ecological variation (Stiner, 1999, 2001; Fiorenza, 2015). Factors other than
ecological variation, such as demographic pressure and available technology, can
determine the proportion of food classes consumed by foragers (Kelly, 1995). One
way to examine the relative influence of these factors is to assess the extent that eco-
geography accounted for variation in food acquisition. Resource choice is a product
of the demands of the individual foraging society, and an increasing population
requires additional energy capture from its territory. Increased energy capture can
be achieved by more intensive use of costlier resources, often with technological

94
specialisation; this model is termed ‘broad spectrum foraging’. Flannery (1969)
envisaged that broad spectrum foraging emerged first at the end of the Pleistocene,
laying a foundation for domestication. Broad spectrum foraging is now thought to
have emerged in intervals that occurred throughout the Upper Palaeolithic in
Eurasia and earlier in Africa (McBrearty and Brooks, 2000). The first appearance of
this pattern has been proposed in the southern Levant and Europe by about 45-30 ka
(Stiner, 1999; Revedin et al., 2010) and North China by the Late Glacial Maximum
(Liu et al., 2013).

As increasing high-resolution methods reevaluate the adoption of broad

spectrum subsistence strategies, Middle Palaeolithic subsistence has received more
attention. The actual appearance of broad spectrum diets may long predate the point
at which they are currently visible in the archaeological record. Some researchers
have pointed to Neanderthal charred legume assemblages from Kebara Cave (63-45
ka) and grass seed phytoliths from Amud Cave (70-55 ka), arguing that the broad
spectrum economy was present already in the Late Middle Palaeolithic (Madella et
al., 2002; Lev et al., 2005). Others have studied starch and phytolith microremains
trapped in dental calculus, and found that Neanderthal dental calculus from sites
such as Spy and Shanidar indicate the use of date palm and grass seeds in the
Levant, and water lily tubers in northern Europe (Henry et al., 2011). Despite these
promising insights into Neanderthal use of plants, these samples are too widespread
in time and space to give reasonable coverage of potential variation in Neanderthal
diets. It is noteworthy that they, these studies tell us little about the longevity of the
Middle Palaeolithic dietary niche. Thus, it is unknown if Neanderthal exploitation of
plant foods broadened over the hundreds of thousands of years they occupied
Eurasia in response to higher populations or milder climates, similar to what is
observed for the Upper Palaeolithic and recent hunter-gatherers, or if variation is
only linked to different ecologies.

To explore the flexibility and stability of Middle Palaeolithic dietary breadth

through environmental variation, we investigated plant consumption as recorded in
dental calculus from environments with varied vegetation and winter and summer
temperatures. We analysed plant microremains trapped in dental calculus from
Neanderthal teeth from five archaeological sites: Vindija (Croatia), Grotta Guattari
(Italy), Grotta Fossellone (Italy), Sima de las Palomas del Cabezo Gordo (Spain) and
Kalamakia (Greece). These samples derive from a variety of regions and biomes
across Europe: The Northern Balkans, and the western, central and eastern
Mediterranean (Fig. 15). We then identified microremains to examine the variety of

95
consumed types or taxa. Once complete we compared this data with previously
published results (Henry et al., 2014; Appendix 7.3) and finally explored if Middle
Palaeolithic dietary breadth varied in different climatic and ecological conditions.
We predicted that if Neanderthal diet was flexible, the number of plant types
represented in the calculus should be greater in warmer, more arboreal
environments. Furthermore, if their population gradually increased dietary breadth,
the number of plant types represented in calculus should be higher at sites that are
more recent.

Fig. 15: Map of western Eurasia with the studied sites indicated.

5.2 Materials and Methods

5.2.1 Sites

Vindija Cave: this cave is situated on the southwest slopes of Kriznjak Peak in
the Hrvatsko Zagorje region of northern Croatia (46°17'N, 16°6'E). Early exploration
of the site began in 1928 with small-scale excavations. Malez and colleagues
conducted large-scale archaeological excavations between 1974-1986 and 1993-1994.

96
These uncovered a complex of 10 m deep strata of 16 layers, with abundant
palaeontological, archaeological and hominin material. A considerable number of
hominin skeletal fragments was found in the cave deposits deriving from five or
more individuals (Karavanić and Smith, 1998). A portion of this material was
Mousterian-associated, and researchers identified the material as coming from Late
Pleistocene Neanderthals due to its less pronounced archaic traits (Smith, Boyd, and
Malez 1985). A radiocarbon date of >45.5 ka cal BP (Krings et al., 2000), and a U/Th
date of a cave bear bone of 50.3 ka cal BP (Wild et al., 2001) have assigned layer G3 to
MIS 3. Direct AMS ultrafiltration dating of hominin remains from layer G1 has
assigned the most recent Neanderthals from this layer to 33,371 ± 399 - 35,382 ± 2224
ka cal BP (Higham et al., 2006). Archaeologists found Neanderthal material mostly in
layers G1 and G3, but also four teeth in Layer F (of which we sampled two: 12.2 and
12.6). There was also modern human material in Layer D (MNI < 10). G3 is
unambiguously Mousterian, while layers G1 and F contain some Aurignacian lithic
material. However, dating and morphological evidence has firmly established the
presence of Neanderthals in these layers, and cryoturbation is likely to have been
responsible for bone displacement (Wolpoff et al., 1981; Higham et al., 2006; Frayer
et al., 2010). Aurignacian lithic typology and early Upper Palaeolithic bone points
are known in layers F and G1. The relatively low density of Aurignacian lithics, the
mixing evident from contradictory dates, and the evidence of Neanderthal traits on
the teeth (Frayer et al., 2010) suggest that the layer F teeth are in fact Neanderthal
remains from layer G, so we feel comfortable including them also in our analyses.
Excavators found red and giant deer (Megaloceros giganteus), elk (Alces alces), and
aurochs (Bos primigenius) in layer G3, chamois (Rupicapra sp.), roe deer (Capreolus
capreolus) and Merck’s rhinoceros (Stephanorhinus sp.) in layer G1 and bison (Bison
sp.), ibex and Merck’s rhinoceros in layer F. Micromammals such as bank voles
(Myodes glareolus) were found in layer G (Mauch Lenardić, 2014). These taxa are
relatively unspecific but generally suggest continental conditions, and fauna such as
roe deer and bank voles suggest at least a proportion of tree cover perhaps as
parkland or riverine mosaics.

Grotta Guattari: this site is one of a complex of caves located in Monte Circeo,
a limestone massif in Lazio, Central Italy (41°14'N, 13°05'E). The site was discovered
in 1939 inadvertently when surface fauna and the remains of one Neanderthal
(Guattari I) in layer G0 were discovered. Later explorations found more
Neanderthals, firstly in a bone scatter (Guattari II) in layer G0, and subsequently in
breccia (Guattari III) at the cave entrance (Sergi, 1954). Of the three Neanderthal

97
Guattari II and Guattari III were sampled. The cave has seven stratigraphic layers
(G0-G5), but G0 is not vertically discrete partially due to carnivore disturbance
(Stiner and Kuhn, 1992). Layers G1-G5 produced lithic artefacts and were deposited
rapidly, but layers G6-G7 are beach deposits that accumulated more slowly (Stiner
and Kuhn, 1992). Researchers identified the hominin remains as morphologically
Neanderthal with a “classic” morphotype, suggesting they date to the Late
Pleistocene (Howell, 1957). Stratigraphically below the fossils are the sequence’s
basal marine-influenced deposits (G7), which are thought to relate to the final high
sea level event of oxygen isotope stage 5a [84-74 ka] (Martinson et al., 1987; Grün
and Stringer, 1991). U-series and electron spin resonance dating of calcite
encrustations on bones and mammal teeth from the stratum that produced Guattari I
and II suggest a date of 60-50 ka, while Guattari III dates to the end of MIS 5, 74-60
ka (Grün and Stringer, 1991; Schwarcz and Schoeninger, 1991). Regional palynology
studies indicate grasslands in cold periods and tree cover in warmer phases (Van
Andel and Tzedakis, 1996; Follieri et al., 1998). A variety of fauna were found on site.
Fauna such as ibex indicate mountainous open habitats, while boar (Sus scrofa) and
roe deer, are thought to indicate tree cover or shrub. Other fauna may represent
either open grasslands/parkland or mixed environments such as Merck’s rhinoceros,
aurochs and mammoth (Elephas antiquus). Extreme cold-adapted species like
reindeer (Rangifer tarandus) or arctic fox (Vulpes lagopus) are absent on coastal sites in
the region, demonstrating the absence of a bitter cold environment (Kuhn, 1991).

Sima de las Palomas del Cabezo Gordo: The site is a karstic vertical cave in
the Permo-Triassic marble hill of Cabezo Gordo overlooking the Mediterranean Sea,
in Torre Pacheco municipality, Murcia, SE Spain (37°47′59″ N, 0°53′45″ W). Much
fossiliferous breccia was extracted from the 18-m-deep entrance shaft by 19th-
century miners and discarded as rubble both on the hillside and inside the cave.
Fortunately, inside the shaft there remained untouched a column of breccia in which
was found a fossil (SP1) of a Neanderthal mandible fused to both maxillae.
Subsequent sieving of rubble and systematic excavations by Walker and Gibert
recovered Neanderthal skeletal elements, Late Pleistocene faunal remains, and
Mousterian Middle Palaeolithic artefacts (Trinkaus and Walker; Walker et al., 2008,
2010, 2011a). The main in-situ archaeological layer has been dated using U-series and
radiocarbon to between roughly 56 and 34 ka (Trinkaus and Walker). Three
articulated Neanderthal skeletons were found in this layer: an adult woman (SP96;
Walker et al., 1999, 2011a, 2012) lying over a child (SP97) below which lay another
adult (SP92). The adult woman SP96 was directly dated using U-series to 54.1 ± 7.7

98
ka (APSLP1) (Walker and Ortega, 2011). Several taxa are typical of the Iberian Late
Pleistocene (Equus caballus, Bos primigenius, Capra pyrenaica, Cervus elaphus, Lynx lynx,
Oryctolagus cuniculus and Testudo hermanni etc.) whereas others occur that rapidly
became extinct at the close of the early Late Pleistocene (Panthera pardus, Crocuta
crocuta, Stephanorhinus sp., Hippopotamus amphibius, Hystrix javanica). Pollen from the
uppermost sediments indicates presence of pines and moisture-dependent
deciduous woodland (which is absent in the region today), and thermophylls
characteristic of southeastern Iberian and North Africa that do not regenerate after
frost (Carrión et al., 2003). Neanderthal teeth with carious lesions have been
identified (Walker et al., 2011b). Teeth sampled for dental calculus come from
excavated sediments except for one (SP50) recovered from the hillside rubble.

Kalamakia: this Middle Palaeolithic site is a cave on the western coast of the
Mani peninsula in the Peloponnese in southern Greece (36°40'43.3"N 22°21'59.3"E).
Archaeologists excavated Kalamakia from 1993 until 2006 (Harvati et al., 2009, 2013).
Chronologists have dated basal deposits with U/Th radiometric dating to the MIS 5c
transgression (109 + 14/−13 ka; De Lumley et al., 1994). Two of the five units
produced substantial Middle Palaeolithic remains (Units III and IV). Excavation
concentrated on Unit IV due to hard breccia in Unit III. Seventeen occupation levels
were identified in the sedimentary deposits of Unit IV. In addition to fauna and
Mousterian lithics, ten hominin teeth, crania and postcranial elements with
diagnostic Neanderthal morphology were found, comprising of at least eight
individuals, three of which we sampled for dental calculus (KAL 3, 5 and 8). Unit
IV’s youngest archaeological level has been dated to >39 ka (Harvati et al., 2013),
placing KAL 5 and KAL 8 between MIS 5a (74 ka) and 39 ka. Excavators uncovered
KAL 3 in Unit III, which overlies 5c beach rock and was truncated by sea
transgressions in MIS 5a. Evidence of other truncations from sea transgressions from
local caves implies that KAL 3 dates to the MIS 5b (Darlas, 2012). Faunal and
palynological studies reveal that prevailing climatic local conditions were mild.
Fallow deer (Dama dama) is particularly common in the assemblages, followed by
ibex, wild boar, red deer, tortoise and some modified seashell. Maquis shrubland
and Mediterranean pre-steppic forest species covered the peninsula (Lebreton et al.,
2008). Extensive avian remains reveal evidence of tree cover in a predominantly
open warm/temperate environment (Roger and Darlas, 2008).

From each site we collected a variety of control samples, including sediments

from the sites, dust on the skeletal material, and samples of the material in which the

99
remains were stored (Appendix table 15, 16, 17, 18, 19). We also tried to sample
dental calculus from the teeth of herbivorous and carnivorous fauna as an additional
control and to explore if Neanderthals, like carnivores, consumed the stomach
contents of herbivores (Buck and Stringer, 2014). Unfortunately, we were only able
to access faunal material from Vindija and Sima de las Palomas del Cabezo Gordo.
These samples included wolf (Canis lupus), which is mostly carnivorous but also
known to consume some plant material; an unspecified feline (c.f. Panthera), which is
nearly strictly carnivorous (Bocherens et al., 2011); and cave bear (Ursus spelaeus),
which had a plant rich diet (Pacher and Stuart, 2009). In addition to the 30
Neanderthal calculus samples from the five sites that we processed for this study,
we also included data from a variety of other northern European, Levantine, and
southern European sites (Appendix 7.3.1) (Salazar-García et al., 2013; Henry et al.,
2014).

In summary, our five sites represent a variety of environmental contexts. They

range from more open temperate environment at Vindija to more Mediterranean
mosaic woodland at Sima de las Palomas del Cabezo Gordo, and from cooler at
Vindija to warmer at Kalamakia. This range reflects the bulk of environments
Neanderthals occupied. We did not try to evenly represent different age classes or
sexes, as often this information is not available.

5.2.2 Dental calculus and control sampling

Neanderthal teeth from each site were examined for deposits of dental
calculus situated on the tooth surface in a cleaned lab of the institution where each
specimen is curated. Deposits of dental calculus were common on teeth examined,
but it was not present on all specimens. We documented the dental calculus deposits
with photography before sampling. We then collected 14 samples of dental calculus
from the Vindija Neanderthal teeth (levels F, G1 and G3), five from the Grotta
Guattari teeth (levels G0), two from the Grotta Fossellone teeth (level 4), six from
Sima de las Palomas del Cabezo Gordo teeth (Upper Cutting level 2 and I), and three
from the Kalamakia teeth (Unit III and Lower IV) (Table 9). Many of the sampled
teeth had a visible band of hard supragingival dental calculus, except the Iberian
teeth that were encrusted in calcium carbonate. In these samples, we therefore took
‘deep’ and ‘shallow’ samples. “Shallow” samples were closer to the surface and
likely to represent the sediment while “deep” ones were more likely calculus.

100
 The sampling surface was gently dry brushed with a disposable toothbrush to
dislodge contaminants at the sampling locations. We then used a dental scalar to
remove small areas of dental calculus onto creased weighing paper underlain by
aluminium foil. The material collected in the paper was then transferred to a
microcentrifuge tube. After sampling, we photographed the teeth and the remaining
unsampled dental calculus. We then transported the samples to the Plant Foods lab
at the Max Planck Institute for Evolutionary Anthropology (MPI-EVA).

To minimize risk of contamination from airborne modern plant material and

lab supplies (Langejans, 2011; Crowther et al., 2014; Henry, 2014), we conducted a
regime of weekly laboratory cleaning. All lab work surfaces were cleaned with hot
water, washed with starch-free soap and with 5 % sodium hydroxide (NaOH). To
assess contamination types, we additionally performed wipe tests before and after
weekly cleaning to quantify starch and other contaminants. Wipe tests retrieved
settled particles of the surface area (74 x 43 cm2) of the laboratory positive-pressure
laminar flow hood used for mounting.

101
 Table 9: Neanderthal dental calculus Grotta Guattari, Grotta Fossellone, Sima de las Palomas del
Cabezo Gordo and Vindija analysed. Dates are ka cal BP.
Sample Site Specimen Tooth Weight (mg)
FON1 Grotta Fossellone Fossellone 3 LL M1 0.067
FON2 Grotta Fossellone Fossellone 3 LL M2 0.1
GTN1 Grotta Guattari Guattari II RL M3 0.654
GTN2 Grotta Guattari Guattari III RL M1 0.871
GTN3 Grotta Guattari Guattari III LL I2 0.654
GTN4 Grotta Guattari Guattari III RL I2 0.258
GTN5 Grotta Guattari Guattari III LL M1 0.289
KAL_3 Kalamakia KAL 3 UL M3 2.866
KAL_5 Kalamakia KAL 5 UR P2 0.05
KAL_8 Kalamakia KAL 8 UR M2 N/A
Vja-13 Vindija 12.1 UR M2 0.393
Vja-14 Vindija 12.2 LR I2 0.046
Vja-16 Vindija 12.4 UR I1 0.046
Vja-17 Vindija 12.5 UR C 0.045
Vja-18 Vindija 12.6 LL C 0.02
Vja-19 Vindija 12.7 LL I2 0.89
Vja-20 Vindija 11.39 LR C 0.446
Vja-21b Vindija 11.39 LR M1 0.408
Vja-21a Vindija 11.39 LR M1 0.502
Vja-24 Vindija 11.45 LL M3 0.672
Vja-26 Vindija 11.46 UL M2 0.865
Vja-51 Vindija 11.4 LL M1 0.19
Vja-54 Vindija 11.4 LL M1 0.046
Vja-55 Vindija 11.4 LL M1 0.085
SP45 Sima de las Palomas del Cabezo Gordo SP45 LR P3 0.08
SP54 Sima de las Palomas del Cabezo Gordo SP54 LR C 0.102
SP78a Sima de las Palomas del Cabezo Gordo SP78 P4 0.415
SP79 Sima de las Palomas del Cabezo Gordo SP79 I1 N/A
SP83 Sima de las Palomas del Cabezo Gordo SP83 LR DM2 0.09
SP84 Sima de las Palomas del Cabezo Gordo SP84 M2 N/A

5.2.3 Sample preparation and mounting

Using standard procedures (Power et al., 2014b) each sample was weighed
and transferred to microcentrifuge tubes while in a clean laminar flow hood at the
Plant Food Group Laboratories at the MPI-EVA. We then ground the samples with a
micropestle in a 1.5 ml Eppendorf microcentrifuge tube containing ~30 µl of a 25 %
glycerine solution to reduce sample loss due to static electricity. The samples were
then centrifuged at 1691 x g (Heraeus MEGAFUGE 16 with TX-400 Swinging Bucket

102
Rotors) for 10 minutes. These samples were mounted on glass slides and examined
under bright field and cross-polarized light on a Zeiss Axioscope microscope at 400 ×
magnification.

5.2.4 Identification and classification

We photographed and described recovered microremains using the

international nomenclature codes (Madella et al., 2005; ICSN, 2011). Phytoliths were
classified into conventional morphotypes, while we developed types to classify other
microremains based on shared morphology. Starches were classified according to
shape, the presence and prominence of lamellae, hilum morphology, formation type
(i.e. simple or compound), cross features, cracks and other surface features. Some
types are unique to a single plant taxon, but in other cases, several types may all
have originated from a single taxon, or one type may be common to several taxa. For
example, several phytolith types (short-cell, bulliform and psilate) may all represent
a single species of grass. When possible, we identified the types to the lowest
taxonomic level possible, usually family or genus (Appendix 7.3). Many categories of
plant foods that could have been important have few or no microremains. These
include lipid-, sugar-, and inulin-rich plants, like olives, walnuts, and Asteraceae tap
roots. Images of all microremains are deposited on the Archaeological Microremain
Database of the Plant Foods in Hominin Dietary Ecology Research Group in Leipzig.

Once we classified the microremains, we calculated ratios that may provide

quantitative information about the assemblage. These included Menhinick’s index, a
richness metric common in ecological studies, which is the ratio of the number of
taxa to the square root of sample size (Magurran, 2004). We used this index to
compare samples to test breadth in each assemblage. We calculated total number of
unique starch and phytoliths types. We also prepared ratios that are phytolith-
specific such as the monocot: dicot phytolith ratio, which may indicate contribution
of grasses, sedges and other monocots versus the contribution of flowering plants;
and the variable: consistent morphology (v/c) phytolith ratio, which indicates taxon.

5.2.5 Palaeotemperature reconstruction

In order to best approximate the climatic conditions of each site, we used

detailed climate simulations for western Eurasia created as part of the Stage 3 Project

103
(Van Andel and Davies, 2003). This project quantified climatic variables during
much of the range of the last glaciation from 59 up to 24 ka, and generated four
regional model simulations: a MIS 3 warm climatic event, a MIS 3 cold climatic
event, the extremely cold Last Glacial Maximum (LGM), and finally a modern
climatic model. These simulations are also created to model conditions in other
periods such as Stage 4 (e.g. Aiello and Wheeler, 1995; Wales, 2012). Unfortunately,
these models cannot account for third order climate fluctuations that occurred
within these phases. However, when each simulation is examined for each
Neanderthal site, we see that the variation in temperatures is driven more by the
latitude and longitude of the site than by the specific climatic period. Therefore,
despite being somewhat coarse-grained, these models allow us to quantify much
temperature variation.

These simulations of temperature can be made more ecologically relevant by

calculating effective temperature, a climatic predictor that evens out yearly
temperature variation. Binford used this powerful measure to explain why recent
forager subsistence varies latitudinally (Bailey 1960; Binford 2001). The necessary
data to calculate effective temperature was unavailable so we developed modified
effective temperature (MET) to adapt effective temperature for available
palaeotemperature data. This differs from effective temperature in that it uses the
mean of the three warmest and three coldest months instead of the warmest and
coldest month. Modified effective temperature is identical to effective temperature in
all other ways. Effective temperature is based on three constants- the minimum
mean temperature (18°C) that supports tropical plant communities (a 365 day
growing season), the minimum temperature (10°C) at the start of the growing season
at the zonal boundary of polar and boreal environments and the minimum
temperature (8°C) at the beginning of the growing season (Binford 1980, 2001).
Modified effective temperature (MET) is as follows

MET={18* MST)-(10*MWT)}/(MST-MWT+8)

where
MST is mean summer temperature (June, July and August)
MWT is mean winter temperature (December, January and February)

The Stage Three Project supplied mean temperature (ºC) 2 m above ground
level from June through August and December through February for each climate

104
simulation. We matched plots of each simulation to the climatic phases covered in
our sample set (Table 10, Table 11), and we collected relevant values from each
simulation plot and then calculated modified effective temperature for each hominin
sample (Table 11; Appendix 7.2).

Table 10: Stage 3 Project simulations used to predict average summer and winter temperatures experienced by
each Neanderthal. Dates are ka cal BP. See Table 11 for the actual predicted temperatures per specimen.

Interval Phase Simulation model used Date

MIS 5e Eemian Interglacial Modern 130-117
MIS 5d Early Glacial Stadial Phase Warm 117-105
MIS 5c Early Glacial Interstadial Phase Warm 105-95
MIS 5b Early Glacial Stadial Phase Warm 94-85
MIS 5a Early Glacial Warm Phase Warm 85-74
MIS 4 Transitional Phase Warm 74-66
MIS 4 First Glacial Maximum Last Glacial Maximum 66-59
MIS 3 Stable Warm Phase Warm 59-44
MIS 3 Transitional Phase Warm 44-37
MIS 3 Early Cold Phase Cold 37-27
MIS 2 Last Glacial Maximum Last Glacial Maximum 27-16

105
Table 11: Palaeoenvironment reconstructions for each specimen used in this study. Tree cover: O=open,
C=closed, M=mixed. P=publication, 1=this study, 2=Henry et al., 2014, 3=Salazar-García et al., 2013. Dates are
ka cal BP.

Specimen Site Date Tree Palaeotemperature MET P

cover Dec- Feb June- Aug
Fossello 3 Grotta Fossellone 70 O -6 16 11.6 1
Guatt II Grotta Guattari 55 O -4 16 11.71 1
Guatt III Grotta Guattari 67 O -6 16 11.6 1
KAL 3 Kalamakia 91 O 4 20 13.33 1
KAL 5 Kalamakia 63 O 4 20 13.33 1
KAL 8 Kalamakia 63 O 4 20 13.33 1
12.1 Vindija 34.3 O -8 20 12.22 1
12.2 Vindija 34.3 O -8 20 12.22 1
12.4 Vindija 34.3 O -8 20 12.22 1
12.5 Vindija 34.3 O -8 20 12.22 1
12.6 Vindija 34.3 O -8 20 12.22 1
12.7 Vindija 34.3 O -8 20 12.22 1
11.39 Vindija 45.5 O -8 20 12.22 1
11.45 Vindija 45.5 O -8 20 12.22 1
11.46 Vindija 45.5 O -8 20 12.22 1
11.4 Vindija 45.5 O -8 20 12.22 1
SP45 Sima de las Palomas del Cabezo Gordo 50 C 4 16 12.4 1
SP50 Sima de las Palomas del Cabezo Gordo 50 C 4 16 12.4 3
SP53 Sima de las Palomas del Cabezo Gordo 50 C 4 16 12.4 3
SP54 Sima de las Palomas del Cabezo Gordo 50 C 4 16 12.4 1
SP58 Sima de las Palomas del Cabezo Gordo 50 C 4 16 12.4 1
SP60 Sima de las Palomas del Cabezo Gordo 50 C 4 16 12.4 3
SP68 Sima de las Palomas del Cabezo Gordo 50 C 4 16 12.4 3
SP74 Sima de las Palomas del Cabezo Gordo 50 C 4 16 12.4 3
SP78 Sima de las Palomas del Cabezo Gordo 50 C 4 16 12.4 1
SP79 Sima de las Palomas del Cabezo Gordo 50 C 4 16 12.4 1
SP83 Sima de las Palomas del Cabezo Gordo 50 C 4 16 12.4 1
SP84 Sima de las Palomas del Cabezo Gordo 50 C 4 16 12.4 1
SP88 Sima de las Palomas del Cabezo Gordo 50 C 4 16 12.4 3
SP100 Sima de las Palomas del Cabezo Gordo 50 C 4 16 12.4 3
Kůlna 1 Kůlna 45 O -8 16 11.5 2
GoyetVII Goyet 40.5 O -8 12 11.5 2
Chapel 1 La Chapelle-aux-Saints 57 O -4 12 10.57 2
Malarn 1 Malarnaud 75 M -4 12 10.57 2
LFI La Ferrassie 39 M 0 12 10.8 2
LFII La Ferrassie 39 M 0 12 10.8 2
Quina V La Quina 64 M -4 12 10.68 2

106
 5.2.6 Palaeoenvironmental reconstruction

In contrast to temperature, we assessed tree cover using all published data on

the habitats that existed at each site. We used investigations of macromammals,
micromammals and pollen that record palaeovegetation at different scales from local
and regional studies to classify each environment. Based on the prevalence of tree
cover we assigned each sample as coming from open, mixed or closed habitats
(Table 11).

5.2.7 Statistical analysis

To explore the relationships among environment, trends in foraging breadth,

and microremains found in our samples and those from previous studies (Salazar-
García et al., 2013; Henry et al., 2014) we fitted an Observational random effect
Poisson model with likelihood ratio tests, using the glmer function of the R package
lme4 (Bates et al., 2013). We chose this Observational model because it is appropriate
for count data, like ours, which is not normally distributed, and instead is skewed
towards zero. We only included those samples that had been weighed prior to
processing, and for which the recovered microremains were assigned to specific
types. If any dental calculus samples produced no microremains, they were included
as zero values. Our full model included modified effective temperature, the
chronological time period in which the age of the Neanderthal lived and tree cover
(open, mixed or closed) as fixed effects. It also included the weight of each dental
calculus sample as model offset to factor in significant differences of sample dental
calculus. We prepared the data by z-transforming age and modified effective
temperature. The site and analyst were treated as random intercept terms. The
weight of the dental calculus sample in mg was included as an offset. An id was
assigned to each observation, and this was also included as a random intercept, thus
removing overdispersion (x2=30.62, df=44, dispersion parameter=0.696). To test the
significance of the full model, it was compared with a null model excluding fixed
effects of modified effective temperature, age of each fossil specimen and tree cover.
Variance inflation factors (VIF) were derived to assess collinearity, from a standard
linear model minus random effects and offsets. Variance inflation factors indicated
collinearity to not be an issue (largest VIF=1.27). We tested model stability by
excluding levels of random effect one by one from the data set, running the full
model and comparing the results with those from the original model that suggest no
highly influential cases. To allow for the possibility of mixing between layers F, G1

107
and G3 in Vindija Cave we built an identical model except that samples from F and
G1 derive from G3. We performed similar checks on this alternative model as the
previous model. We removed overdispersion on this model (x2=32.90, df=0.89,
dispersion parameter=0.748) and ensured VIF was not an issue (largest VIF=1.371).

5.3 Results

5.3.1 Contamination controls

Vindija Cave: We collected some samples of faunal calculus, as well as

adhesives used to hold Vindija tooth 11.39 (Appendix 7.3.4). Bear, wolf and felid
samples from Vindija yielded 81 plant microremains, but these were
disproportionally more common on bear than wolf samples, consistent with the
expected diets of these species (Pacher and Stuart, 2009).

38 phytoliths were found in bear and wolf samples. Multicellular polyhedrons

were abundant in one bear sample, reflecting rich consumption of dicot fruit and
leaves. Multicellular polyhedrons are rare or absent in most wolf samples except Vja-
12-31. We found a small number of dicot phytoliths in a few other wolf samples.
Present-day wolves consume plant matter, and plants may comprise up to 40 % of
their food intake in certain seasons (Meriggi et al., 1991). European wolves especially
favour fruit, but wolves may also consume plants in stomach contents or
intentionally consume grass to smooth digestion or ease parasites (Murie, 1944;
Stahler et al., 2006).

Starches were particularly uncommon in these fauna samples. Most were

small nondiagnostic types. A Triticeae grass aggregate was found in a wolf control
sample (Vja-30). Total numbers of starches found and the number of starches per
milligram were lower than in Neanderthal samples (Appendix 7.3.4). Furthermore,
the faunal samples appear to share similar starch types (1-7 types), while the
Neanderthal calculus had more varied starches (1-15 types). Two control samples of
mandible adhesive revealed 56 contaminant starches but nearly all of these were
heavily damaged potato starch. These starch are morphological distinct from those
in the Neanderthal dental calculus samples (Appendix table 15, Appendix table 16,
Appendix table 17, Appendix table 18, Appendix table 19).

Grotta Guattari and Grotta Fossellone: We took a variety of control samples,

though not all preferred control types (e.g. faunal teeth) were available. Most
controls were samples of adhesives used to bond bone, or washes of distilled water
108
taken from the surfaces of the sampled mandibles. These contamination assays
produced no or few microremains, and where microremains were found they
showed a narrow range of types (Fig. 16; Appendix 7.3.4). We found that these
samples contained few types of starch, and contaminating grains appeared distinctly
fresh and usually occurred as starch aggregates unlike more damaged and isolated
starch in dental calculus samples. A Triticeae grass seed starch aggregate was found
in controls 2e and Fon3. None of this type of aggregates were found in hominin
samples.

Sima de las Palomas del Cabezo Gordo: In addition to controls (non-worked

stone from archaeological strata, carnivore dental calculus and packing cotton)
published in Salazar-García et al., 2013, we sampled carnivore dental calculus and
sediment found attached to hominin teeth. One sediment sample produced a single
isolated subspherical starch. These results show a very low rate of background
starch and phytoliths.

Kalamakia: We did not have access to any contamination control materials for
Kalamakia.

160

140
Total starch & phytoliths types/mg

120

100

80

60

40

20

0
Balkan Balkan fauna Balkan Cen & East Western West
Neanderthals controls Mediterranean Mediterranean Mediterranean
Neanderthals Neanderthals fauna

Fig. 16: Starch and phytoliths from Neanderthal calculus, fauna calculus and controls show that Neanderthal
dental calculus samples show a distinct signal indicating they reflect hominin diet.

109
 5.3.2 Dental calculus microremain assemblages and dietary breadth

Vindija Cave: We collected calculus from six isolated teeth and five in situ
teeth (Table 9). Isolated teeth included a right second molar, a lower second incisor,
upper first incisor, upper canine, lower canine, and lower second incisor. Our
sample of in situ teeth included a lower canine, a lower third molar, an upper second
molar, and a lower first molar. Microremains were recovered in all Neanderthal
dental calculus samples but there was major variation in the numbers and classes
present. The plant microremain assemblages found on the Vindija samples is
considerably more diverse than what was reported in the previous studies of
Neanderthal calculus (Hardy et al., 2012; Henry et al., 2012, 2014; Salazar-García et
al., 2013).

229 microremains were found in Vindija dental calculus samples, including 87

starch microremains (Appendix 7.3.4). 15 starches displayed a lenticular cross-
section, circular or subcircular plane view, a hilum exhibiting a thin line, and
distinctive surface dimples and lamellae, clearly representing starches from Triticeae
grass seeds (Fig. 5). Although grass leaf microremains may arise from non-edible
resources such as bedding, this seems unlikely to be the case for grass seeds.

Two of the starches are likely to derive from a legume based on their
characteristics: circular, oval, ovoid shape, the presence of lamellae, and the
characteristic longitudinal cleft fissure. We have observed these traits in peas (Pisum
sp.), vetches (Vicia sp.), and sweet peas/vetchlings (Lathyrus sp.). Three other starches
(Fig. 17) displayed the size, highly faceted surface and polyhedral shape consistent
with those of starches from hard endosperm (Eliasson and Larsson, 1993). Plants that
produce this starch morphology include nuts, hard seeds, seeds from grasses not in
the Triticeae tribe, and seeds of sedges like Schoenoplectus. Two starches from
underground storage organs were evident from large elongated shapes and highly
eccentric polarisation crosses. None of these legume, hard endosperm, or
underground storage organ starches had specific enough morphological
characteristics to identify them to a lower taxonomic category. The remaining
starches fall into nine groupings, probably reflecting several taxa, but due to starch
damage, redundant types and a limited reference collection, they cannot be
identified. Five starch types also found in Neanderthal samples were also found in
cave bear samples, but these were nondiagnostic types.

We recovered 91 phytoliths from the Vindija dental calculus samples

(Appendix 7.3.4). Thirty-two of the 91 were long cell morphotypes, which are

110
common in the leaves of monocots. However, because monocots produce more
phytoliths than dicots per gramme (Tsartsidou et al., 2007), they are more visible in
the archaeological record. Phytolith production between the two categories varies
from 80:1 to 20:1 (Tsartsidou et al., 2007). Ratios of monocot to dicot in our sample of
Vindija Neanderthal dental calculus vary from 5:1 to 0.67:1, which suggests an
abundance of dicot types such as fruits, nuts and leaves rather than grasses and
sedges.

Twenty-five spores were also found, representing approximately five types of

fungus. However, these are nondiagnostic and could represent mushroom-bearing
higher fungi or lower fungi such as moulds. Pollen was rare and only one ‎Betulaceae
pollen was found. Ten unsilicified plant tissue fragments were recovered, two
reflecting grass and one an unspecific monocot, but others were indeterminate.

Grotta Guattari and Grotta Fossellone: We examined the calculus from the
right lower third molar of Grotta Guattari II and the lower first molars (right and
left) and lower second incisors (right and left) of Grotta Guattari III. Calculus
samples from the five teeth from Grotta Guattari produced high numbers of
microremains and high levels of diversity per mg. A total of 151 microremains were
found in the dental calculus of the five teeth (Appendix 7.3.4).

Starch grains were found on four of the five teeth and totalled to 69 grains. Six
starches found in an amyloplast cell were elongate ovoid in plane-view and oval in
cross-section, with an eccentric polarisation cross, all characteristics matching Lilium
type starches (Fig. 3). One starch clearly represented a Triticeae grass seed starch.
Further evidence of grass use is evident from intact grass leaf tissue found in one
sample. The other detected starches represented five unknown types.

Thirty-nine phytoliths were recovered, 31 of which originated in monocot

tissue and eight from dicot plants. Nine short cell rondel phytoliths were identified.
One phytolith was a multicellular epidermal jigsaw morphotype, indicating dicot
leafy or fruit matter. We also note the presence of a tracheid vessel, which is another
dicot marker.

Other microremains were numerous. Ten spores were observed, some of

which exhibited features that enabled us to identify them as coming from bracken
(Pteridium sp.). We also noted the presence of spores from Nigrospora sp. We also
identified other spores such as fusiform spores, indicative of boletoid fungi. Many
bolete fungi are edible and widely consumed, while Nigrospora is a diverse genus of

111
fungi that are mostly agents of decay. Five pollen grains were found including two
Betulaceae pollens. In total 14 other cellular plant tissue fragments were noted,
including vascular bundles reflecting plants that entered the mouth. Also recovered
were a number of stellate hairs and a pennate diatom.

Fig. 17: Mosaic of microremains and comparative modern reference plant matter. Each scale bar represents
10 μm. (a) Starch from Vindija Neanderthal identified as Triticeae under bright field and (b) cross polarized
light, (c) a reference Triticeae starch (Triticum turgidum sp.) under bright field and (d) cross polarized, (e)
Amyloplast with several ovoid starches resembling Lilium bulb starches under bright field and (f) cross
polarized light, (g) reference Lilium sp. bulb starches under bright field and (h) cross polarized light, (i)
polyhedral starch under bright field and (j) cross polarized light, (k) Pteridium sp. spore, (l) diatom embedded
in calculus, (m) fragment of grass leaf, (n) triporate Betulaceae pollen, (o) Unsilicified tracheid plant tissue.

Grotta Fossellone: We sampled dental calculus from the left lower first molar
and second molar of Grotta Fossellone III. Eleven starches were found in the two

112
Grotta Fossellone dental calculus samples. These comprised of indeterminate
starches that cannot yet be matched to reference material. Only one phytolith was
found in the assemblage: a rondel phytolith from a grass. Additionally, one piece of
monocot and one piece of unidentified plant tissue were found.

Sima de las Palomas del Cabezo Gordo: For this study, we sampled dental
calculus from six hominin teeth, including a lower third premolar, lower canine,
lower deciduous second molar, a lower forth premolar, an upper first incisor, and a
lower first molar. We found relatively few microremains in these samples, reflecting
the very small amount of dental calculus in each sample. We recovered only five
starches and phytoliths, and one diatom. None could be identified to plant taxon.

Kalamakia: We sampled dental calculus from three Kalamakia teeth - an

upper molar (KAL 3), an upper fourth premolar (KAL 5) and an upper molar (KAL
8). Only five starch grains were found on the three teeth. Two phytoliths were also
found: one from a grass and one from a non-monocotyledon plant. Sixteen possible
calcium oxalate forms were found. Calcium oxalate represents consumed plant
matter, but it is readily soluble and occurs in most plants, and is therefore not
assignable to taxon. Lastly, we found one fragmented sponge spicule. This last
microremain likely entered the mouth through accidental consumption.

5.3.3 Dietary flexibility and dietary niche stability

We predicted that if the breadth of Neanderthal plant use was driven by

ecological conditions, then the number of consumed types should be influenced by
temperature and tree cover. We produced a Menhinick’s index comparison of all
available samples, including all previously published data and the new samples
from this study. Although there is no distinct trend among Neanderthals from
different periods or chronologies (Fig. 18; Appendix 7.3.4), there is a possible
curvilinear relationship, with the Menhinick’s index /mg increasing with
temperature until a peak is reached, at which point the index drops again. It is
possible this pattern reflects the degradation of starches in the warmest
environments (Langejans, 2010).

We condensed the list of samples to include only those with documented

weights (Wt column, Appendix 7.3.4). We then used an observational random effect
Poisson model to test dietary breadth patterns (described above). We find no
relationship between the number of microremain types found in calculus and the

113
chronological age or environmental conditions of the sample, even when accounting
for the effects of variation between tree cover, sites, analyst, age, and weight of the
dental calculus sample (Appendix table 20). More specifically, an increase in
temperature did not lead to an increase in the number of types represented in dental
calculus and younger sites did not show an increase in the number of types
represented in dental calculus (x2=5.148, df=4, P=0.273; Appendix fig. 3; Appendix
table 20). Even in the alternative model, which assumed bones in Vindija Cave layer
G1 are older than thought and derive from G3, there was still no relationship
(x2=2.683, df=4, P=0.612; Appendix fig. 3; Appendix table 20).

12

10
Menhinick's index (/mg)

0
10.57 10.67 10.8 11.5 11.6 11.71 12.22 12.4 13.33
Modified Effective Temperature (MET)

Fig. 18: A Menhinick’s index of types of starch and phytolith from Neanderthal dental calculus shows that
warmer climates are not associated with increased diversity. Samples are from Neanderthal remains
presented in this study and Salazar-García, et al., 2013 and Henry et al., 2014. Each dash represents an
individual sample.

5.4 Discussion

Microscopy revealed starch and phytoliths in most samples but many

samples were highly variable. Fig. 16 and Appendix fig. 3 shows that many dental
calculus samples from Grotta Fossellone, Sima de las Palomas del Cabezo Gordo and
Kalamakia yielded few microremains. Previous work that established baselines with
chimpanzee (Power et al., 2015b) and living human (Leonard et al., 2015)
populations indicates that this stochastic pattern is normal. These studies also

114
emphasize that we have not recovered information on the majority of consumed
plants when using this approach.

Our previous work with chimpanzees indicated that age influences the
microremain record, with older individuals having more microremains and more
plant types represented. Though we do not have precise age estimates for the
individuals in this study, all of the teeth were from adults. Furthermore, we have no
reason to expect that any site was biased towards individuals from one particular
age class. Taken together, these data suggest that the differences among the sites do
not reflect simple age differences within our sample.

By using this metric of dietary breadth we were able to show that

Neanderthals in warmer environments who had better access to arboreal edible
resources might not have used a broader range of plant foods than Neanderthals
from colder environments, and in some cases they even show less diversity than cool
climate ones. This picture could arise by methodological limitations of this approach,
since it is possible that plant remains such as starches are underrepresented in
samples from warmer environments due to worsened taphonomic conditions (Smith
et al., 2001). However, the phytoliths follow a similar pattern in our results, despite
being insensitive to temperature, suggesting that the observed pattern in our study
samples is due to dietary, not post-depositional, trends. Our results on microremain
diversity do not negate occlusal dental wear findings that link tree cover to plant
use, as occlusal wear only approximates classes of the total diet and not its
composition. The availability of Pleistocene plant foods, however, likely reflects
forest type (Mediterranean or Boreal) far more than tree cover alone. Open and
mixed environments have less primary biomass than closed canopy environments,
but they may offer significantly more edible plant biomass, as much of the biomass
in forests consists of tree trunks, and is thus unavailable to hominin consumers
(Odum, 1975). Pleistocene aridity may also have encouraged plant use; amongst
recent foragers at a given latitude plant consumption usually increased in more open
environments, largely because aquatic animal foods are less available in these dryer
habitats (Keeley, 1992).

The plants used indicate how Neanderthals sourced nutrition from their
environment. We find evidence of the use of grass seeds, true lily tubers, legumes
and other starchy plants that leave no taxon-attributable types. Other microremains
types included pollen and spores. Spores from Guattari III suggest interaction with
fungi but these spores are too rare to ascertain the presence of deliberate interaction

115
with fungi such as the consumption of mushrooms (Power et al., 2015a). Not all
recovered microremains reflect intentionally consumed food. Recovery of Betulaceae
pollen and bracken spores may highlight use of birch, or hazel and bracken, but as
these particles are excellent dispersers, they probably simply reflect characteristics of
the airborne suspensions and aerosols in the Pleistocene airborne environment.

Some of the types we were able to identify also tell us about Neanderthal
dietary behaviour. In particular, many of the microremains come from low-ranked
foods, like grass seeds and tubers (Simms, 1985). Grass seeds are widely used by
recent foragers in warm and cool environments (Lothrop, 1928; Simms, 1985; Harlan,
1989; Brand-Miller and Holt, 1998). Grass seeds used at Vindija and at Guattari
demonstrate an investment in a low-rank plant food in cool habitats of the northern
Balkans and coastal Italy. The use of grass seeds is often linked to terminal
Pleistocene Southwest Asian foragers invested in broad spectrum diets because grass
seeds are usually costly to harvest and prepare for consumption (Simms, 1985). On
the other hand, there is abundant evidence that groups like the Vindija Neanderthals
were big game hunters and that energetic contribution from plants is not likely to
have rivalled meat. Middle Palaeolithic foragers probably only used grass seed as a
limited component of the broader plant diet as this resource offers limited
nutritional return (Simms, 1985). This is the same pattern observed in Upper
Palaeolithic human foragers of Southwest Asia where grass seed use is most
prominent (Savard et al., 2006; Rosen, 2010).

Overall, there is no indication that Neanderthals gradually used a more

diverse array of plants, despite some evidence of a modest increase in population
from 70 ka onwards (Foley and Lahr, 2003; Van Andel and Davies, 2003; Speth and
Clark, 2006). The possible absence of a chronological trend in vegetal dietary breadth
agrees with the lack of a trend in their predation niche prior to 55 ka. Yet dental
calculus may hint that Neanderthal vegetal dietary breadth diverged from the
narrow spectrum hunting economy. While the exploitation of small, fast and hard-
to-catch game necessitated a costly increase in technology, some plants can be
harvested and processed without technological investment. Although this may
contradict conventional expectations of glacial period foragers in Central Europe, the
cold temperatures of Pleistocene Eurasia may mislead us on the ecological
productivity of this region. The pattern is better explained by decoupling seed and
nut use from the dietary expectations of broad spectrum diets. Low intensity use of
plants outside broad spectrum diets is possible (Hockett and Haws, 2003; Revedin et

116
al., 2010). Although an expanding plant food niche may be a sign of demographic
packing population increase, its presence need not signify a total investment in
complex foraging/broad spectrum foraging if such plant exploitation was possible
without costly plant harvesting and processing technology (Hockett and Haws,
2003). Non-intensive use of these plants was possible with the technology available
to Neanderthals.

Neanderthals could have reduced their processing costs by making use of

caches of USOs and seeds, such as rodent stores (underground winter food stores),
and by choosing to harvest the plants during seasons when they were easiest to
prepare. The raiding of rodent stores requires little technology, though it often
requires considerable ecological knowledge (Jones, 2009). For example, Siberian
peoples raided rodent stores to obtain Lilium tubers all year round (Ståhlberg and
Svanberg, 2010, 2012), but they had to be able to discern edible tubers from toxic
USOs. Neanderthals ecological knowledge may have also been useful for the
consumption of grass seeds. As Neanderthals exhibit no evidence of plant
processing or food storage, we propose Neanderthals collected these seeds without
laborious and expensive processing costs. One of the few ways this is possible is by
plucking green grain from spikelets before they ripen and harden (Rosner, 2011).
Unlike ripe grain, green grain requires no grinding or pulverising and may be
consumed once dehusked, which can be done by hand. Green grain starch grains are
smaller than those of ripe grain but they share most morphological characteristics
and are likely to be identified as Triticeae with our methodology (Evers, 1971). Green
grain is a resource that is only available in a narrow window before the grain ripens
into a hard dry grain (Rosner, 2011). This collection of green grain would be
suggestive of precisely seasonally organized Neanderthal foraging. Unfortunately,
there is insufficient data to reconstruct a seasonal round of plant food gathering, but
gathering during at least the spring is evident

5.5 Conclusions

The dental calculus microremain assemblages present strong evidence of

Neanderthal use of various plants as foods, and complement our understanding of
Neanderthal subsistence. This suggests that plant-harvesting strategies existed
alongside their hunting economy. Plant foods were likely valued for their micro- or
macronutrient profiles rather than caloric energy alone. Hominin physiology limits
the total dietary protein intake, impeding an absolute reliance on protein-rich foods

117
such as terrestrial mammals lean meats (Cordain et al., 2000; Speth, 2010; Hockett,
2012). Recent foragers have avoided the effects of protein overconsumption by
incorporating other macronutrients in diet. Foragers often source animal fat as the
preeminent strategy for offsetting risk of protein poisoning (Speth and Spielmann,
1983; Cordain et al., 2000). However, animal fat from a diet of terrestrial ungulates
may have been insufficient. Triticeae, Fabaceae and Liliaceae offer rich sources of
carbohydrates that may have offset the problems of lean protein consumption.

The incorporation of diverse plant foods including those with low- or middle–
ranking returns into the human diet probably predates Neanderthal diets, has a long
history in the human lineage, and is likely that such diets persisted throughout
hominin evolution mediated by energetic ecological necessity and labour
availability. Similarly, resource depletion-driven subsistence changes may have
occurred at many points in hominin evolution. Indeed it is observed elsewhere in
Homininae, in present day chimpanzees, where increases in chimpanzee
populations have been linked to increased use of low ranked prey (Watts and
Mitani, 2015).

Regarding Neanderthal subsistence, we find no evidence of variation

throughout the final 60,000 years they occupied Eurasia. Our model also finds no
indication that plant use was confined to certain parts of their range. Surprisingly we
did not find that a more diverse range of types were consumed in southern areas.
Although this may suggest dietary inflexibility, it could also reflect relatively
unchanging strategies, stable thanks to their success. While past research has
revealed unappreciated variability in Neanderthal animal food use (Stiner, 1994;
Speth and Clark, 2006), as a whole animal food provision centred on hunting of large
and medium-sized game and thus Neanderthals exhibit lower intensity of diverse
resources than early modern counterparts in Eurasia (Richards et al., 2000, 2001). A
large- and medium-sized game hunting economy supplemented with plant foods
may have evolved as a specialisation strategy in response to Eurasian environments
(Stiner, 2013).

118
 6

Discussion: A pathway for reconstructing Neanderthal

Dietary Ecology

This dissertation opened with the great conundrum of Neanderthal ecology.

How can Neanderthal diet be defined and its apparent distinctiveness explained
when surviving evidence is so incomplete? By now, it is observable that most of the
surviving evidence illustrates diets that are unrealistically carnivorous. Instead of
relying solely on evidence that is intractable, it is necessary to extrapolate diet from
indirect lines of evidence, ethnographic analogy and alternative methodologies such
as dental calculus analysis, which is the focus of this work. The work presented
above aimed to improve the method of analysing plant microremains trapped in
dental calculus as a means to reconstruct diet, and then apply the lessons learned
from methodological exploration to a study of the diets of Neanderthals. Each of the
individual research projects contributed to the greater whole, while simultaneously
raising its own issues and concerns; some of which will be discussed below.

6.1 Developments in dental calculus analysis and dietary reconstructions

The publication of applications of dental calculus analysis has greatly

outpaced the progress of its theoretically and methodologically driven justification.
There is a shortage of data on the extent that dental calculus analysis methodologies
retrieve dietary debris that is present in dental calculus (Hardy et al., 2009; Henry,
2014; Leonard et al., 2015). This problem implied that a central part of the
dissertation needed to bridge the increasing use of dental calculus for dietary
inferences and the characteristics of dental calculus that allow it to trap foods;
necessitating. One outcome is that this dissertation built a practical framework that
allows reliable analysis of the dental calculus record in spite of its variability.

Improving the methodology available for dental calculus research has

broadened the role that dental calculus can play in reconstructing the life history of
Palaeolithic hominins. These validation studies enhance our ability to make
inferences about Neanderthal diet because they provided information on the
limitations of dietary interpretation from microremains preserved in dental calculus.

119
The methods appraisal undertaken with chimpanzee calculus from the Taï Forest in
Côte d'Ivoire, and human dental calculus from the Chalcolithic site of Camino del
Molino in southeastern Iberia (Power et al., 2014b), revealed insights to the
microenvironments and structures in dental calculus that preserve food debris and
other markers of life history. High-resolution analysis identified numerous
microremains in-situ on the outer extents of dental calculus matrices. The
identification of these remains shows how dental calculus preserves complex and
diverse assemblages of plant, animal, and fungal microremains, all of which are
useful for studying life histories of our ancestors. The comparative approach
established that microremain types retrieved are a product of chosen sample
preparation and microscopy technique. In particular, I noted that a combined SEM-
OM approach was most useful for researchers interested in a variety of
microremains, but that OM alone was a swifter and better technology for identifying
the taxonomic origins of starch grains, which can be important in reconstructing
diet. The study showed that researchers must customise their analytical technique to
the desired suite of microremains (Power et al., 2014b). A sequential workflow
translated these findings into an effective means of revealing underrepresented
types relating to consumed food and water as well as the local environment (See 3.4).

The dissertation has built upon this exploration by investigating the

representativeness of dental calculus as a dietary record. My work accomplished this
by closely matching recorded diet to the plant microremain assemblages in dental
calculus, using wild chimpanzees from the Taï Forest as a validation population
(Power et al., 2015b). The project was able to compare over two decades of dietary
data from 128 chimpanzees observed by primatologists to the selected group of 24
chimpanzees (only 24 of the available skeletons were from known individuals with
documented age and sex). After building an extensive reference collection of Taï
Chimpanzee plant foods, my study predicted how microremains in dental calculus
would record diet. I then cross-validated the calculus record with the dietary
observations. In the process of carrying out the analysis of this reference sample, the
research pioneered machine learning identification of botanical remains in dental
calculus. This technique matched each microremain to the most likely source plant
genus. In addition, it yielded a score of confidence to each identified microremain.
This reduced some of the subjectivity and biases commonly present in conventional
identification approaches.

With this comparison, the analysis quantified the resolution of the dental
dietary record. It showed that there is a relationship between the number of starches,

120
phytoliths and other microremains with diet. However, diagnostic-starch-producing
plants represented only 25 % of total feeding time in the dietary records, whereas
diagnostic-phytolith-producing plants represented just 2 % of feeding time. Many of
the plants that chimpanzees eat produce few or unidentifiable microremains.
However, a more extensive study of the Taï chimpanzee diet would plausibly be
able to cover a substantially larger portion of diet as the time restrictions on my
project prevented examination of many food plants. A large portion of unidentified
diet likely was reflected in the dental calculus by non-starch and non-phytolith
microremains, some of which may once become identifiable. Only 49.56% of
microremains (starches and phytoliths) in the dental calculus were analysed and the
remaining (plant fragments, calcium oxalate, and pollen) were not. Microremain
patterns showed that starch, phytoliths, and other microremains accumulate in
dental calculus over the lifetime of an individual, yet it was apparent that this
process is subject to a number of factors that are hard to account for. A number of
non-dietary factors may influence microremains in chimpanzee calculus, such as sex,
but this remains to be confirmed. My statistical analysis found that the frequency of
phytolith a genus correctly predicted the time spent consuming that plant genus. Yet
with starches, the proportion of each genus poorly predicted its actual dietary
importance, as it is probably thwarted by uncontrolled taphonomic factors. The
record provides only a minimum estimate of the number of plant types consumed.
However, even starch microremains provided details on specific chimpanzee
resource choices that are unavailable with all other methods. It can record the
presence of key resources important for specific behaviours, in my case the first
introduction of foods as a result of weaning, and the consumption of hard to open
nuts, which are linked to social attributes including learned behaviour.

When applied to the plant microremain record more generally, the results of
these studies suggest several patterns. First, the relative quality of plant
microremains depends on the microremain type. Phytoliths are far more likely to
survive taphonomic processes and have no known preservation limit owning to
their molecular make-up. Unlike phytoliths, starches are not robust and microbial
breakdown often underrepresents these microremains. Starches are often damaged,
and degrade beyond a certain age (Collins and Copeland, 2011), although this starch
half-life is not yet known. It is reasonable that local conditions may preserve starch
and phytolith contrastingly (Langejans, 2010). It is clear from our Chalcolithic group
and Neanderthal samples that starches tend to be more abundant than phytoliths in
human calculus. This reflects the fact that starch, unlike phytoliths, is a sought after
nutrient among human groups.

121
 Microremain analysis on dental calculus is one of the few methods to explore
the taxonomy of foods that have entered the mouth. Documenting specific types of
exploited plants is a window on how important plant foods were to Neanderthals. If
taphonomic conditions have not degraded calculus dietary data, microremains can
preserve some of the lifetime diversity of vegetal diet. Diversity may be a useful
metric of the vegetal component to diet in itself. It is also valuable for assessing the
ability of foragers to identify diet- related ecological knowledge. This is the
knowledge of the useful properties of specific plant taxa and the proficiency in
extracting in varied complex sequences according to the correct growth cycle stage
(Jones, 2009).

There would be no ambiguity in stating that calculus analysis will not detect
the majority of Neanderthal foods. Fortunately, dental calculus can preserve details
about food plants correspondently absent from macrobotanical studies. If used in
isolation dental calculus analysis would build a misleading picture of Neanderthal
subsistence strategies. A dental calculus approach is suited to multi-disciplinary
research using dental wear, isotopic analysis, ideally with large collections of fossil
individuals along with macrobotanical screening on archaeological sites.

6.2 New information on Neanderthal diets revealed by microremains

The Neanderthal microremain assemblages showed evidence of the use of

grass seeds, lilies, legumes, and other starchy plants that do not leave taxon-
attributable types. Of these plants, legumes and grass seeds have also been identified
in Middle Palaeolithic macrobotanical assemblages at Kebara Cave. At this site,
legumes were overwhelmingly dominant, whereas grass seed was only a minor
component, the opposite pattern to the one observed across the samples I examined.
Likewise, the plant taxa recovered from Middle Palaeolithic sites found at Douara
Cave, Gorham’s Cave, Mas-des-Caves and Rabutz did not overlap with my results.
However, all of these assemblages contain few taxa and low number of individual
seeds and thus are not representative. Furthermore, most of the charred and
desiccated botanical remains so far found on these sites are lipid-rich nuts (e.g. olive
and hazelnut) that produce no or few starch or phytoliths. Thus, these plants are
unlikely to be represented in dental calculus. In addition, my results agree with
sediment-based phytolith studies at Amud Cave, which also reported significant use
of grass seeds (2.5.5). This study was based on dendritic phytoliths and thus reflects
deposited grass seed husks rather than edible endosperm matter represented by

122
starch. Grass seed husks identified at sites such Amud Cave may be accidental
inclusions, but when combined with the starch record found in calculus in this and
other studies, which is unlikely to be accidentally introduced, these results provide
strong support to the idea that Neanderthals consumed grass seeds. This is one
example of how different types of microremains reveal varied and complementary
information.

Our rigorous contamination controls and weekly tests confirmed that the
large majority of microremains in the Neanderthal samples were endogenous,
however, a few remains, particularly the fibres (numerous in Vindija dental calculus;
Appendix table 15), could be contaminants. The microremains that I was able to
distinguish from contamination and identify as ancient markers of life history
included starch grains, phytoliths, plant and fungi spores, other plant tissues,
diatoms and mineral particles. These were not present in all samples and some
whole sites such as Kalamakia exhibited very few microremains. I did not try to infer
the total vegetal contribution to diet because my chimpanzee findings demonstrate
that it is not yet possible to interpret the total dietary contribution of plants from the
total microremain numbers. The feasibility of this is not established because the
dental calculus record appears too stochastic. This might be further confounded if
Neanderthals cooked starch-containing foods, or if they removed phytoliths from
food plants before use. Although my study of chimpanzees used a population
approach, this is not possible in these Palaeolithic sample as very few Neanderthal
remains are available and due to the impossibility of knowing if remains are
contemporaneous. Our studied Neanderthals are a collection of individuals rather
than a population sample. Neanderthal diet may have exhibited considerable
variation according to sex and age like recent hunter-gatherers, but this cannot to
examined as so few samples are available.

The resources with evidence of consumption include plants whose roles to

Late Pleistocene foragers are usually overlooked by archaeologists. Diet breadth
models and the ethno-historic record contextualise the significance of these food
plants. Foods such as grass seeds and legumes are likely to have been low-rank for
central European Neanderthals (Simms, 1985; Kelly, 1995; Savard et al., 2006). In
more recent societies, these foods are often hallmarks of commitment to low-rank
plant foods because of the high amount of processing used to access their energy
dense nutrients For example, as such plants grew in importance in terminal
Pleistocene societies, processing of these plants began to dominate life and restrict
mobility (Wright, 1994; Molleson, 2000). One problem of these foods is they occur as

123
small packages of nutrients, which are dwarfed by nuts and underground storage
organs. Yet these foods offer the advantage of occurring predictably annually, while
nuts often occur periodically in three to five year cycles, which occur synchronously
over geographic regions (Vander Wall, 2001). Section 5.4 proposed Neanderthals
collected these plants without laborious and expensive processing costs. This
purported collection of green grain would be suggestive of highly seasonal
Neanderthal foraging. Even if harvested outside of this window, the use of these
resources indicate there was a seasonal round that allowed exploitation of resources
unavailable for most of the year. One alterative way to source high search cost foods
is to raid rodent caches of tubers and seeds. These caches may store many kilograms
of edible and nutritious plant foods throughout the year (Nabhan, 2009; Ståhlberg
and Svanberg, 2010).

In mild and humid parts of Neanderthal range, a broader range of plant foods
including leafy greens, drupes and berries as well as USOs occurred for a large part
of the year but elsewhere they were more restricted. In cool dry regions some plant
foods such as USOs are always present in the environment, for recent foragers they
become less accessible in winter. In northern winters, plant foods become locked in
frozen soils, buried in snow, or trapped under lake ice. However, the animal food
supply also diminishes over winter as the condition of prey deteriorates as they
expend body fat. These factors result in long winters depleting energy supply in
northern foraging diets. The extent to which plant foods became inaccessible to
Neanderthals during winter months is unclear, and if they used them to bridge this
period of scarcity. When plant foods took a marginal role for acquiring energy, they
may have been sourced for micronutrients, as well as macronutrients as emergency
fallback foods, when high-ranked winter resources failed. Hunting medium and
large game unusually produces an irregular food supply, and due to their reliance
on hunting Neanderthals may have suffered from being in an ecologically precarious
position. Regular plant consumption possibly alleviated some of this risk but the
regularity that fallback foods may have been required should not be understated.
The archaeological record suggests plant use and other fallback foods were
insufficient to avoid frequent local extinctions (Hublin and Roebroeks, 2009;
Snodgrass and Leonard, 2009).

The range of consumed plants is unlikely to be explained by opportunistic

Neanderthal foragers who only occasionally used plants. It is notable that this study
shows that Neanderthals used plant foods (grass seeds and legumes) outside of the
traditional period of food scarcity in the northern hemisphere (late winter to early

124
spring). This pattern signifies regular use of plant foods. Isotopic and microwear
demonstrate Neanderthals were predominantly consuming meat but microremains
indicate that Neanderthals were not the Pleistocene equivalents of the near
carnivorous recent Arctic foragers. However, due to the absence of habitats
equivalent to Pleistocene Eurasia, there is no ethnographic analogy for Neanderthals
(Stringer et al., 2000; Stewart, 2005; Zimov et al., 2012). This is especially troublesome
for Neanderthals in open environments because the rarity of ethnographic analogies
compounds the lack of a recent habitat parallels (Kelly, 1995). This issue may also
influence or distort sister studies of Neanderthal diet using recent foragers as
analogies (e.g. dental wear). Anthropologists documented few temperate grassland
foragers aside from horse cultures of the American Plains. However, in recent
forager societies occupation of cooler climates correlates with reliance on aquatic
resources and food storage (Kelly, 1995; Cordain et al., 2000). The scarcity of
Neanderthal examples for either of these dietary resources and techniques implies a
unique dietary niche with potential oversupply of protein and undersupply of
certain fatty acids in certain seasons (Section 2.4.2). The risk of these problems was
mediated by plant gathering. Carbohydrates offered energy to offset or defray the
potential costs of excessive reliance on muscle issue, which could otherwise lead to
protein poisoning (Cordain et al., 2000; Speth, 2010). It is also possible that essential
fatty acids vital for development were available in certain wild plant foods
(Simopoulos, 2004). Plant consumption must have been crucial element of
Neanderthal ecology.

6.3 The response of Neanderthal subsistence to varied environments

Neanderthals are the evolutionary outcome of the reproductive isolation from

African populations in the environments of northern Eurasia for hundreds of
thousands of years. In this period, Neanderthals and their ancestors persisted
through severe climatic change due to glacial cycles. These cooling and drying
events transformed Eurasian fauna and flora, affecting the habitability of the region
itself. Neanderthals reacted to events in various ways, and in much of their
distribution, their range contracted (See 2.2.2). Given that, Neanderthals inhabited a
plethora of different environments from arid steppe to boreal forest to warm coastal
Mediterranean woodland, it is possible that they possessed sufficient cultural
knowledge to accommodate glacial cycle environment change. Microwear analysts
have proposed that Neanderthals altered their diet in periods of climatic fluctuation
(El Zaatari et al., 2016).
125
 My dissertation attempted to identify the dietary signal of glacial phases and
climate by testing microremain diversity. Contrary to expectations, the model of
plant consumption revealed that plant use breadth appears relatively static across
their range in time and space (See 5.4). This possibly argues that Neanderthal plant
use shows a limit to their behavioural flexibility. Not all plants produce distinct
microremains, therefore some dietary variation in plant use may not be visible with
this method. In modern environments, edible wild plants provide subtly different
suites of macro and micronutrients across varying habitats. If we assume, as is
suggested by evidence, that Neanderthals were not solely seeking energy, then they
may have foraged for different plant foods in the different habitats they occupied.
However, the model does indicate local adaptations were limited. Perhaps this
conservatism mirrors the limited cultural evolution in Neanderthal technology
(Stiner, 2013). The failure of Neanderthals to acquire more regional foraging
repertoires is not necessarily maladaptive, or reflective of their cognitive abilities. It
is clear they were a successful species, having dominated western Eurasia for
hundreds of thousands of years without interruption. Simply put, they had a very
different suite of behaviours than those seen among modern or recent historical
humans.

6.4 Archaic hominin diet and social structure

The choice of resources used by this hominin has implications for

Neanderthal socioeconomics. The specialisation of labour by sex in recent foraging
cultures is deeply intertwined with how edible plants are available (See section
2.3.1). Previously researchers have taken Neanderthal hunting and lack of highly
specialised technological investment to infer far less division of labour compared
with recent foragers (Kuhn and Stiner, 2006). The breadth of plant use indicated by
past dental calculus analysis has been interpreted as suggestive of a sexual division
of labour closer to recent foragers than previously thought (Henry, 2010).
Unfortunately, the dissertation data cannot confirm this level of sociality, but the
observed diversity of plant exploitation does agree with the idea that a specialisation
of labour was present. It is probable that, if it existed, a plant gathering specialisation
was adopted by females due to nursing. Sexual division of labour and food sharing
are likely to have been essential for pregnant and nursing females to overcome
seasonal shortfalls of food supply.

126
 6.5 The use of palaeoecological models for inferring subsistence

In this dissertation, I made inferences on the environments Neanderthals

occupied in order to model their plant use. The parameters used were
palaeotemperature and dominant local vegetation. By using the outputs of the Stage
Three Project, the dissertation had a rich insight into climate but the
palaeotemperature simulations used have a limited resolution, which is inevitable
given that climate is an enormously variable system. Combining this with the
thousands of years of climate change in Late Pleistocene Europe adds further
complexity. Even if the simulations are accurate, they are averages of second order
events that lasted thousands of years. If a Neanderthal specimen was from a climatic
phase affected by an anomalous third order temperature variation, our estimates
may be incorrect. Furthermore, I had to rely on matching the map graduations from
each temperature simulation to the coordinates of each archaeological site. This
approach potentially lost fine detail of each simulation that is not visible on the
project output maps. If this approach were to be used in the future, it would be
improved by developing an algorithm to match each archaeological site to each
simulation. Other useful available climatic parameters are annual temperature
range, precipitation and snow depth and these may have refined the model if sample
size was large enough to permit their conclusion.

The project also calculated if the localities of these sites were forested, mixed
or devoid of trees. In contrast to the regional level temperature reconstruction, this
vegetation reconstruction used proxies (pollen and mammal assemblages) for local,
or mesoscale environmental reconstruction. Undoubtedly, more regional level or
megascale environmental models would enrich the interpretative power of the
study. However, for this extra variable to allow further dissecting of Neanderthal
habitats a greater calculus sample size would be required.

6.6 Future directions

In spite of the rise of knowledge on Middle Palaeolithic dietary ecology, diet

in this period remains poorly comprehended. An exploration of eco-geography and
plant consumption is required to address this. Increasing scale by increasing sample
size will make it possible to catalogue dietary variation. The expansion of dental
calculus, dental wear, and stable isotopic research is needed, particularly in the large
spans of their northern, southern and eastern range that remain underrepresented in

127
dietary studies. If larger sample sizes are available, scientists may be able to estimate
the majority of starch- and phytolith-rich plants they consumed. If data becomes
available on the other Pleistocene archaic hominins that existed contemporary to
Neanderthals in Eurasia and Africa it would be possible to contextualise
Neanderthal resource use.

Future research will also need to interpret implications of plant use in

Pleistocene ecologies. This requires exploring the energy returns of plant and animal
foods in Pleistocene environments with models of Neanderthal demographics.
Nutritional returns and processing costs would allow us to analyse food choice with
diet breadth models. It would allow us to predict the rank of specific plant taxa and
consider if Neanderthals were targeting bulk energy, specific nutrients or other
traits. In no case is there a robust sequence of the steps of use of different taxa - a
chaîne opératoire - that follows gathering through processing to consumption. Some
of the plants identified in this dissertation could be explored with historical and
experimental archaeological studies (Haws, 2004). Studies of the costs and returns of
harvesting, processing and consumption of identified plant foods with Middle
Palaeolithic technology would assist this. Yet ultimately, experimental archaeology
is chained by the limited span of the ethnographic record and Neanderthals may
have used these plants in ways unintelligible with the ethnographic record.

Most of what is known about the diet of this Pleistocene relative of our
species only describes their last 40,000 years. Researchers have inadequate details
about plant foods and subsistence over the approximate preceding 150,000 years
they occupied Eurasia. How stable their plant food niche was during their earlier
history is a fundamental question. Presumably, diet varied between the warm
interglacial and cold glacial phases. Information is also needed to assess if the
ancestors of Neanderthals who colonised Eurasia used the same taxa as
Neanderthals. Did adapting to Eurasia involve a changing reliance on plant foods?
Perhaps inherited gathering strategies receded and new hunting strategies emerged
as these hominins moved to colder climes. The higher environmental productivity in
southern regions tells us that these hominins probably used more plants than
Neanderthals (Kelly, 1995). However, this may not be the case if they originated in a
more arid climate as less plant biomass is available in arid areas (Kelly, 1995).

This dissertation highlights some of the problems in the current techniques

used to study microremains in dental calculus. It is recommended that further
investigations be carried out to continue to investigate these issues. By performing
an appraisal of methodology in the discipline, my research has built a platform of
128
approaches that anthropological and archaeological scientists can use to advance
dental calculus for dietary research. Scientists can now test how variability in the
formation history of dental calculus in other groups and regions may influence
assemblages of plant microremains. Now researchers know more clearly that dental
calculus is highly variable in composition and in abundance, so analysts need to
account for this in the interpretation of diet.

However, researchers still know little about the stimulus and tempo of
biomineralisation of dental calculus in the mouth. The community needs to consider
the process of biomineralisation and its ability to preserve long-term dental calculus
dietary histories. This may be possible using a more multifaceted approach that
combines methods used in this work with Raman and micro-Fourier Transform
Infrared spectrometry. Momentum is already substantially increasing in avenues
such as mass spectrometry, genetic and proteomic approaches. Further research that
integrates these technologies on contemporary reference samples and as well as
ancient sample could overcome some of these unaddressed obstacles (Charlier et al.,
2010; Adler et al., 2013; Warinner et al., 2014). One recent study (Warinner et al.,
2014) has attempted to bring several of these techniques together but only on an
archaeological sample where there are major uncontrolled confounding effects.
Researchers have not yet combined these techniques on a reference group with a
known diet. Such an attempt would offer to resolve questions on the temporal span
represented by calculus. These approaches may directly examine diet too. In some
cases, these techniques, such as genetic and lipid studies, could be used on dental
calculus to find traces of food evident from microremains and cross validate
microremain evidence of Neanderthal diet.

One of the observations of the project is that particles in dental calculus

cannot simply reflect diet. Microremains in dental calculus must also reflect the
environment that the individual occupied. Some microremain types included pollen
and spores, which are often found in airborne suspensions and aerosols. This topic is
of concern for the question of respiratory health of Pleistocene foragers as some of
these airborne particles are environmental irritants. Little is known about air or other
types of pollution that Neanderthals must have endured (Hardy et al., 2015b; Monge
et al., 2015). Further studies may be able to quantify air-carried microremains,
perhaps as a way to extrapolate Pleistocene suspensions and aerosols.

All dental calculus dietary studies are dependent on the sedimentary

processes that alter and breakdown the archaeological record (taphonomy). These
processes are increasingly relevant as investigators attempt dental calculus analysis
129
on hominin fossils. Extending dental calculus research to earlier Middle Palaeolithic
hominins samples will be valuable for acquiring information about the diets of these
archaic humans. Researchers have reported starches on sites as early 420–200 ka at
Qesem Cave (Israel). Yet there is no hard data on the chronological and
environmental limits of preservation of degradable plant debris in dental calculus.
The study of taphonomy of dental calculus remains in its infancy. Most dental
calculus studies rely on major assumptions about taphonomy, and downplay its
potential to influence results. Yet it seems unwise to underestimate the complex and
variable impact that taphonomy may have. Chemical, fungal and microbial
processes all play a part in breaking down food debris particles in dental calculus.
With starches, there is also the potential risk of spontaneous decay over long periods
of time (Collins and Copeland, 2011). To account for taphonomy future research
could prioritise assessing if dental calculus is sealed from external agents. Much
work is needed to combine elemental composition assays using thin sections of
archaeological dental calculus to clarify the factors contributing to preservation.

6.7 Conclusions

Until relatively recently hunting dominated Palaeolithic literature, but plant

foraging was either ignored or received minimal attention. This occurred even
though plants were almost certainly the primary food source for most of the history
of the Hominidae family (Butterworth et al., 2016). Fortunately, the emergence of
new methodologies has awoken interest addressing this discrepancy. Particularly
dental calculus analysis, the focus of this research, has encouraged examination of
Palaeolithic and especially Neanderthal foraging.

In evaluating the role of dental calculus analysis for reconstructing

Neanderthal foraging, this dissertation with high-resolution approaches helped to
explain how dietary microremains are preserved in dental calculus. However, since
microremains in dental calculus are not useful without contextualisation, Chapter
Two quantified the reliability of dietary phytoliths and starches in a Taï Chimpanzee
population. The representativeness of these assemblages varies considerably and
sometimes according to factors that are challenging to control, but the findings do
establish that these assemblages can mirror diet. Although no technique offers a
high-precision reconstruction of diet composition, dental calculus can make a unique
contribution to dietary studies when it is combined with other techniques. These
insights affirm the value of dental calculus for gaining an insight into ancient diet.

130
 Past studies have highlighted that the use of plant resources is not a hallmark
of modern humans, and that they may be common feature of Neanderthal
subsistence (Hardy et al., 2012; Henry et al., 2014). Not only do the samples in this
study support these findings, many also exhibit a diversity in microremain types
that exceeds that of past Neanderthal dental calculus studies (Henry et al., 2011,
2014; Hardy et al., 2012). This was an unexpected pattern as it was undocumented in
other calculus studies. Yet, my research was able to detect this diversity thanks to the
observations from the Camino del Molino and the Taï Chimpanzee samples.

Evidence of plant consumption from many different lines of evidence has left
a powerful impression that Neanderthal diets cannot be defined by hunting alone.
The identification of plant use has provoked a wave of discussion on Palaeolithic
dietary ecology. The new direction that researchers have taken has kindled a new
paradigm in dietary ecology that is far more aware of the potential breadth of
hominin diets and how frequently plants play an important role (Barton et al., 1999;
Lev et al., 2005; Henry et al., 2011; Sołtysiak, 2012). The findings of my dissertation
hint at how widespread plant use may have been. However, there has been a lag in
translating this evidence into sufficiently nuanced subsistence models. Some
research has interpreted any level of Neanderthal plant use as suggestive of broad
spectrum foraging, but this model fits empirical data poorly. The broad spectrum
foraging concept emerged to describe specific characteristics of pre-agricultural
societies in southwest Asia, and other regions where agriculture emerged. Attempts
to identify broad spectrum foraging in other regions (Jones, 2016) have been in in
vain. Arguably, this model is unsuited to foragers who occupied habitats with
differing climate and technology. In cooler climates, the broad-spectrum framework
may not be readily applicable due to habitat and other differences. Narrow spectrum
foraging, with a marked emphasis on large game hunting as well as fishing,
continued by some inland northern foragers late into the Holocene (e.g. Yesner,
1989). Researchers must find a more appropriate set of concepts for foragers in
cooler climates where trajectories of change were expressed in different ways.

Ethnographic evidence and dental wear studies unequivocally convey that

plants were more important in southern regions (2.3.3; 2.5.7) but this eco-geographic
variation is undetectable in the series of isotopic studies so far conducted. Now with
the dental calculus data, we can infer that eco-geographic variation is also not
apparent with the range of resources used. Although my work’s diet model is not a
full model of dietary breadth, it documents a degree of plant diet breadth. The
results show that the number of plant foods consumed did not vary detectably

131
between climates (5.3.6). Although this finding is unexpected, it parallels the relative
dietary homogeneity between different regions that is indicated by multiple isotopic
studies (2.5.8) and the slow pace of technological development in the Middle
Palaeolithic. During the Upper Palaeolithic, plant use increased over time (El Zaatari
and Hublin, 2014). However, Neanderthal plant use appears homogenous through
the tens of thousands of years represented by our sample, reinforcing the picture of
Neanderthal dietary staticity.

My thesis argues that this evidence of Neanderthal plant use suggests plants
were an essential feature of subsistence but does not contradict evidence of a
Neanderthal economy centred on medium and large game. Nor does this imply that
Neanderthal dietary ecology was necessarily identical or similar to that of the
modern humans who colonised Eurasia during the Upper Palaeolithic. The
variability of Neanderthal diets is clearly less than modern human diets in Eurasia
(Richards et al., 2000, 2001). Yet it is possible that both Neanderthals and modern
humans had optimal diets from a diet breadth perspective, which maximised the
nutritional opportunity available with their respective technology. However,
Neanderthals differ from moderns by exhibiting lower levels of variability.
Although to a certain extent Neanderthal diet may have been rigid from region to
region, this does not imply a shortfall in Neanderthal adaptability. Neanderthal
dietary ecology was specialised to the specific fauna and flora conditions to Eurasia.
The reliance on terrestrial mammals and plant foods should be seen as interaction
with the hyper-arid Pleistocene climates. Further work will find the temporal and
geographic boundaries of their unique adaptation to Eurasia.

132
References
Abraham J, Grenón M, Sánchez HJ, Pérez C, Barrea R. 2005. A case study of elemental and
structural composition of dental calculus during several stages of maturation using
SRXRF. J Biomed Mater Res A 75:623–8.

Ackerknecht EH. 1948. Medicine and disease among Eskimos. Eskimo, Ciba Symp 10:916–
21.

Addy M, Shellis RP. 2006. Interaction between attrition, abrasion and erosion in tooth wear.
In: Lussi A, editor. Dental Erosion: From Diagnosis to Therapy. Vol. 20. Basel:
Monographs in Oral Science, Karger Publishers. p 17–31.

Adler CJ, Dobney K, Weyrich LS, Kaidonis J, Walker AW, Haak W, Bradshaw CJA,
Townsend G, Sołtysiak A, Alt KW, Parkhill J, Cooper A. 2013. Sequencing ancient
calcified dental plaque shows changes in oral microbiota with dietary shifts of the
Neolithic and Industrial revolutions. Nat Genet 45:450–455.

Aiello LC, Wheeler P. 1995. The Expensive-Tissue Hypothesis: The Brain and the Digestive
System in Human and Primate Evolution. Curr Anthropol 36:199–221.

Albert RM, Lavi O, Estroff L, Weiner S, Tsatskin A, Ronen A, Lev-Yadun S. 1999. Mode of
Occupation of Tabun Cave, Mt Carmel, Israel During the Mousterian Period: A Study
of the Sediments and Phytoliths. J Archaeol Sci 26:1249–1260.

Albert RM, Ruíz JA, Sans A. 2016. PhytCore ODB: A new tool to improve efficiency in the
management and exchange of information on phytoliths. J Archaeol Sci 68:98–105.

Albert RM, Weiner S, Bar-Yosef O, Meignen L. 2000. Phytoliths in the Middle Palaeolithic
Deposits of Kebara Cave, Mt Carmel, Israel: Study of the Plant Materials used for Fuel
and Other Purposes. J Archaeol Sci 27:931–947.

Ambrose SH. 1990. Preparation and characterization of bone and tooth collagen for isotopic
analysis. J Archaeol Sci 17:431–451.

Van Andel TH, Davies W. 2003. Neanderthals and modern humans in the European
landscape during the last glaciation: archaeological results of the Stage 3 Project.
Cambridge: McDonald Institute for Archaeological Research.

Van Andel TH, Davies W, Weninger B. 2004. The human presence in Europe during the last
glacial period I: human migrations and the changing climate. In: Neanderthals and
modern human in the European Landscape during the Last Glaciation. Cambridge:
McDonald Insititute Monographs. p 31–56.

Van Andel TH, Tzedakis PC. 1996. Palaeolithic landscapes of Europe and environs, 150,000-
25,000 years ago: An overview. Quat Sci Rev 15:481–500.

Anderson JP. 1939. Plants Used by the Eskimo of the Northern Bering Sea and Arctic
Regions of Alaska. Am J Bot 26:714–716.

133
Antunes MT, Santinho-Cunha A. 1992. Neanderthalian remains from Figueira Brava cave,
Portugal. Geobios 25:681–692.

Arensburg B. 1996. Ancient dental calculus and diet. Hum Evol 11:139–145.

Armitage PL. 1975. The extraction and identification of opal phytoliths from the teeth of
ungulates. J Archaeol Sci 2:450–455.

Armstrong RA, McGehee R. 1980. Competitive exclusion. Am Nat 115:151–170.

Baer DJ, Gebauer SK, Novotny JA. 2012. Measured energy value of pistachios in the human
diet. Br J Nutr 107:120–125.

Bailey HP. 1960. A method of determining the warmth and temperateness of climate. Geogr
Ann 42:1–16.

Bank RA, Hettema EH, Muijs MA, Pals G, Arwert F, Boomsma DI, Pronk JC. 1992. Variation
in gene copy number and polymorphism of the human salivary amylase isoenzyme
system in Caucasians. Hum Genet 89:213–222.

Barton H, Torrence R. 2015. Cooking up recipes for ancient starch: assessing current
methodologies and looking to the future. J Archaeol Sci 56:194–201.

Barton RNE, Currant AP, Fernández-Jalvo Y, Finlayson JC, Goldberg P, MacPhail R, Pettitt
PB, Stringer CB. 1999. Gibraltar Neanderthals and results of recent excavations in
Gorham’s, Vanguard and Ibex Caves. Antiquity 73:13–23.

Bar-Yosef O. 1998. The Natufian culture in the Levant, threshold to the origins of agriculture.
Evol Anthropol Issues, News, Rev 6:159–177.

Bates D, Maechler M, Bolker B. 2013. lme4: Linear mixed-effects models using S4 classes. R
Packag version 0999999-2:999999.

Beaumont J, Geber J, Powers N, Wilson A, Lee-Thorp JA, Montgomery J. 2013. Victims and
survivors: stable isotopes used to identify migrants from the Great Irish Famine to 19th
century London. Am J Phys Anthropol 150:87–98.

de Beaune SA. 1993. Nonflint stone tools of the Early Upper Paleolithic. In: Knecht H, Pike-
Tay A, White R, editors. Before Lascaux: The Complex Record of the Early Upper
Paleolithic. Boca Raton: CRC Press. p 163–191.

de Beaune SA. 2004. The invention of technology. Curr Anthropol 45:139–162.

Beauval C, Lacrampe-Cuyaubère F, Maureille B, Trinkaus E. 2006. Direct radiocarbon dating

and stable isotopes of the neandertal femur from Les Rochers-de-Villeneuve (Lussac-
les-Châteaux, Vienne). Bull Mémoires la Société d’anthropologie Paris 18:35–42.

Behringer V, Borchers C, Deschner T, Möstl E, Selzer D, Hohmann G. 2013. Measurements of

Salivary Alpha Amylase and Salivary Cortisol in Hominoid Primates Reveal Within-
Species Consistency and Between-Species Differences. PLoS One 8.

Belitz HD, Grosch W, Schieberle P. 2009. Food chemistry. Berlin: Springer-Verlag.

134
Berbesque JC, Marlowe FW. 2009. Sex differences in food preferences of Hadza hunter-
gatherers. Evol Psychol 7:601–616.

Bergström J. 1999. Tobacco smoking and supragingival dental calculus. J Clin Periodontol
26:541–547.

Bermúdez de Castro JM. 1997. A Hominid from the Lower Pleistocene of Atapuerca, Spain:
Possible Ancestor to Neandertals and Modern Humans. Science (80- ) 276:1392–1395.

Binford LR. 1968. Post-Pleistocene adaptations. In: Binford SR, Binford LR, editors. New
Perspectives in Archaeology. Chicago: Aldine. p 314–341.

Binford LR. 1980. Willow Smoke and Dogs’ Tails: Hunter-Gatherer Settlement Systems and
Archaeological Site Formation. Am Antiq 45:4–20.

Binford LR. 1985. Human ancestors: Changing views of their behavior. J Anthropol Archaeol
4:292–327.

Binford LR. 2001. Constructing frames of reference: an analytical method for archaeological
theory building using hunter-gatherer and environmental data sets. Berkeley:
University of California Press.

Bird DW, O’Connell JF. 2006. Behavioral ecology and archaeology. J Archaeol Res 14:143–
188.

Blasco R, Fernández Peris J. 2009. Middle Pleistocene bird consumption at Level XI of

Bolomor Cave (Valencia, Spain). J Archaeol Sci 36:2213–2223.

Blasco R, Fernández Peris J. 2012. A uniquely broad spectrum diet during the Middle
Pleistocene at Bolomor Cave (Valencia, Spain). Quat Int 252:16–31.

Blasco R, Rosell J, Arsuaga JL, Bermúdez de Castro JM, Carbonell E. 2010. The hunted
hunter: the capture of a lion (Panthera leo fossilis) at the Gran Dolina site, Sierra de
Atapuerca, Spain. J Archaeol Sci 37:2051–2060.

Blasco R, Rosell J, Fernández Peris J, Arsuaga JL, Bermúdez de Castro JM, Carbonell E. 2013.
Environmental availability, behavioural diversity and diet: a zooarchaeological
approach from the TD10-1 sublevel of Gran Dolina (Sierra de Atapuerca, Burgos, Spain)
and Bolomor Cave (Valencia, Spain). Quat Sci Rev 70:124–144.

Blondiaux J, Charlier P. 2008. Palaeocytology in skeletal remains: microscopic examination

of putrefaction fluid deposits and dental calculus of skeletal remains from French
archaeological sites. Int J Osteoarchaeol 18:1–10.

Bocherens H. 2009. Neanderthal Dietary Habits: Review of the Isotopic Evidence. In: Hublin
J-J, Richards MP, editors. The Evolution of Hominin Diets: Integrating Approaches to
the Study of Palaeolithic Subsistence. Dordrecht: Springer Netherlands. p 241–250.

Bocherens H, Drucker DG, Billiou D, Patou-Mathis M, Vandermeersch B. 2005. Isotopic

evidence for diet and subsistence pattern of the Saint-Césaire I Neanderthal: review and
use of a multi-source mixing model. J Hum Evol 49:71–87.

135
Bocherens H, Drucker DG, Bonjean D, Bridault A, Conard NJ, Cupillard C, Germonpré M,
Höneisen M, Münzel SC, Napierala H, Patou-Mathis M, Stephan E, Uerpmann H-P,
Ziegler R. 2011. Isotopic evidence for dietary ecology of cave lion (Panthera spelaea) in
North-Western Europe: Prey choice, competition and implications for extinction. Quat
Int 245:249–261.

Boesch C. 1997. Evidence for dominant wild female chimpanzees investing more in sons.
Anim Behav 54:811–815.

Boesch C. 2012. Wild Cultures: A Comparison between Chimpanzee and Human Cultures.
Cambridge, UK: Cambridge University Press.

Boesch C, Boesch H. 1983. Optimisation of Nut-Cracking with Natural Hammers by Wild

Chimpanzees. Behaviour 86:265–286.

Boesch C, Boesch H. 1984. Possible causes of sex differences in the use of natural hammers
by wild chimpanzees. J Hum Evol 13:415–440.

Boesch C, Boesch-Achermann H. 2000. The Chimpanzees of the Taï Forest: Behavioural

Ecology and Evolution. Oxford, UK: Oxford University Press.

Boesch C, Marchesi P, Marchesi N, Fruth B, Joulian F. 1994. Is nut cracking in wild

chimpanzees a cultural behaviour? J Hum Evol 26:325–338.

Boule M. 1889. La caverne de Malarnaud, près Montseron (Ariège). Bull la Société Philomath
Paris 8th 1:83–86.

Boule M. 1911. L’Homme fossile de La Chapelle-aux-Saints. Ann Paléontologie 6:106–172.

Bouyssonie A, Bouyssonie J, Bardon L. 1913. La station Moustérienne de la “Bouffia”

Bonneval a la Chapelle-aux-Saints. Anthropologie 24:609–640.

Boyadjian CHC. 2012. Análise e identificação de microvestígios vegetais de cálculo dentário

para a reconstrução de dieta sambaquieira: estudo de caso de Jabuticabeira II, SC.

Boyadjian CHC, Eggers S, Reinhard K. 2007. Dental wash: a problematic method for
extracting microfossils from teeth. J Archaeol Sci 34:1622–1628.

Brand-Miller JC, Holt SH. 1998. Australian aboriginal plant foods: a consideration of their
nutritional composition and health implications. Nutr Res Rev 11:5–23.

Breiman L. 2001. Random forests. Mach Learn 45:5–32.

Brown JK. 1970. A Note on the Division of Labor by Sex. Am Anthropol 72:1073–1078.

Brown K, Fa DA, Finlayson G, Finlayson C. 2011. Small game and marine resource
exploitation by Neanderthals: the evidence from Gibraltar. In: Trekking the Shore. New
York, NY: Springer. p 247–268.

Buck LT, Stringer CB. 2014. Having the stomach for it: a contribution to Neanderthal diets?
Quat Sci Rev 96:161–167.

136
Buckley S, Usai D, Jakob T, Radini A, Hardy K. 2014. Dental calculus reveals unique insights
into food items, cooking and plant processing in prehistoric central Sudan. PLoS One
9:e100808.

Bull ID, Matthew J. Lockheart, Mohamed M. Elhmmali, David J. Roberts, Richard P.

Evershed. 2002. , Evershed RP (2002) The Origin of Faeces by Means of Biomarker
Detection. Environ Int - Google Scholar. Environ Int 27:647–654.

Butterworth PJ, Ellis PR, Wollstonecroft M. 2016. Why protein is not enough: the role of
plants and plant processing in delivering the dietary requirements of modern and early
Homo. In: Hardy K, Kubiak-Martens L, editors. Wild harvest: plants in the Hominin
and pre-agrarian human worlds. Oxbow. p 31–55.

Butterworth PJ, Warren FJ, Ellis PR. 2011. Human α-amylase and starch digestion: An
interesting marriage. Starch - Stärke 63:395–405.

Cabanes D, Mallol C, Expósito I, Baena J. 2010. Phytolith evidence for hearths and beds in
the late Mousterian occupations of Esquilleu cave (Cantabria, Spain). J Archaeol Sci
37:2947–2957.

Cabanes D, Weiner S, Shahack-Gross R. 2011. Stability of phytoliths in the archaeological

record: a dissolution study of modern and fossil phytoliths. J Archaeol Sci 38:2480–2490.

Capitan L, Peyrony D. 1912a. Station préhistorique de La Ferrassie, commune de Savignac-

du-Bugue (Dordogne). Rev Anthropol:29–50.

Capitan L, Peyrony D. 1912b. Station préhistorique de La Ferrassie, commune de Savignac-

du-Bugue (Dordogne). Rev Anthropol:76–99.

Capitan L, Peyrony D. 1912c. Trois nouveaux squelettes humains fossiles. Rev

Anthropol:439–442.

Carpenter D, Dhar S, Mitchell LM, Fu B, Tyson J, Shwan NAA, Yang F, Thomas MG,
Armour JAL. 2015. Obesity, starch digestion and amylase: association between copy
number variants at human salivary (AMY1) and pancreatic (AMY2) amylase genes.
Hum Mol Genet 24:3472–80.

Carrión JS, Yll EI, Walker MJ, Legaz AJ, Chain C, Lopez A. 2003. Glacial refugia of
temperate, Mediterranean and Ibero-North African flora in south-eastern Spain: new
evidence from cave pollen at two Neanderthal man sites. Glob Ecol Biogeogr 12:119–
129.

Carter JA. 1999. Late devonian, permian and triassic phytoliths from antarctica.
Micropaleontology 45:56–61.

Castellano S, Parra G, Sánchez-Quinto FA, Racimo F, Kuhlwilm M, Kircher M, Sawyer S, Fu

Q, Heinze A, Nickel B, Dabney J, Siebauer M, White L, Burbano HA, Renaud G, Stenzel
U, Lalueza-Fox C, de la Rasilla M, Rosas A, Rudan P, Brajković D, Kucan Ž, Gušic I,
Shunkov M V., Derevianko AP, Viola B, Meyer M, Kelso J, Andrés AM, Pääbo S. 2014.
Patterns of coding variation in the complete exomes of three Neandertals. Proc Natl

137
 Acad Sci U S A 111:6666–71.

Cerling TE, Mbua E, Kirera FM, Manthi FK, Grine FE, Leakey MG, Sponheimer M, Uno KT.
2011. Diet of Paranthropus boisei in the early Pleistocene of East Africa. Proc Natl Acad
Sci U S A 108:9337–41.

Charlier P, Huynh-Charlier I, Munoz O, Billard M, Brun L, de la Grandmaison GL. 2010. The

microscopic (optical and SEM) examination of dental calculus deposits (DCD). Potential
interest in forensic anthropology of a bio-archaeological method. Leg Med 12:163–71.

Christophersen KM, Pedersen PO. 1939. Investigations into dental conditions in the neolithic
period and in the bronze age in Denmark. Dent Rec (London) 59:575.

Churchill SE. 2014. Thin on the Ground: Neandertal Biology, Archeology and Ecology.
Oxford, UK: John Wiley & Sons.

Churchill SE, Rhodes JA. 2006. How strong were the Neandertals? Leverage and
muscularity at the shoulder and elbow in mousterian foragers. Period Biol 108:457–470.

Codron D, Lee-Thorp JA, Sponheimer M, de Ruiter D j., Codron J. 2008. What Insights Can
Baboon Feeding Ecology Provide for Early Hominin Niche Differentiation? Int J
Primatol 29:757–772.

Collins MJ, Copeland L. 2011. Ancient starch: Cooked or just old? Proc Natl Acad Sci U S A
108:E145, author reply E146.

Colliot G, Anderson Patricia C., Bonnet N. 1997. Preliminary classification of phytolith

shapes using computerized image analysis and pattern recognition. In: The State-of-
the-Art of Phytoliths in Soils and Plants. . p 275.

Constantino PJ, Lee JJ-W, Chai H, Zipfel B, Ziscovici C, Lawn BR, Lucas PW. 2010. Tooth
chipping can reveal the diet and bite forces of fossil hominins. Biol Lett 6:826–9.

Cordain L, Miller JB, Eaton SB, Mann N, Holt SH, Speth JD. 2000. Plant-animal subsistence
ratios and macronutrient energy estimations in worldwide hunter-gatherer diets. Am J
Clin Nutr 71:682–92.

Cortés-Sánchez M, Morales-Muñiz A, Simón-Vallejo MD, Lozano-Francisco MC, Vera-

Peláez JL, Finlayson C, Rodríguez-Vidal J, Delgado-Huertas A, Jiménez-Espejo FJ,
Martínez-Ruiz F, Martínez-Aguirre MA, Pascual-Granged AJ, Bergadà-Zapata MM,
Gibaja-Bao JF, Riquelme-Cantal JA, López-Sáez JA, Rodrigo-Gámiz M, Sakai S, Sugisaki
S, Finlayson G, Fa DA, Bicho NF. 2011. Earliest known use of marine resources by
Neanderthals. PLoS One 6:e24026.

Coster ACF, Field JH. 2015. What starch grain is that? – A geometric morphometric
approach to determining plant species origin. J Archaeol Sci 58:9–25.

Crawford MA, Broadhurst CL, Ghebremeskel K, Sinclair AJ, Saugstad LF, Schmidt WF,
Cunnane SC, Galli C, Holmsen H. 2008. The role of docosahexaenoic and arachidonic
acids as determinants of evolution and hominid brain development. In: Tsukamoto K,
Kawamura T, Takeuchi T, T. D. Beard J, M. J. Kaiser, editors. Fisheries for Global

138
 Welfare and Environment, 5th World Fisheries Congress. Tokyo: Terrapub. p 57–76.

Crevecoeur I, Bayle P, Rougier H, Maureille B, Higham T, van der Plicht J, De Clerck N,

Semal P. 2010. The Spy VI child: A newly discovered Neandertal infant. J Hum Evol
59:641–656.

Cromton AW, Hiiemae K. 1970. Molar occlusion and mandibular movements during
occlusion in the American opossum, Didelphis marsupialis L. Zool J Linn Soc 49:21–47.

Crowther A, Haslam M, Oakden N, Walde D, Mercader J. 2014. Documenting contamination

in ancient starch laboratories. J Archaeol Sci 49:90–104.

Damen JJM, Ten Cate JM. 1989. The effect of silicic acid on calcium phosphate precipitation.
J Dent Res 68:1355–1359.

Darlas A. 2012. Geomorphologic evolution and occupation of the caves of the western Mani
peninsula during the Upper Pleistocene and the Holocene. Proc 5th Symp Hell Soc
Archaeom:237–253.

Debénath A, Jelinek AJ. 1998. Nouvelles fouilles à La Quina (Charente) : Résultats

préliminaires. Gall Préhistoire 40:29–74.

Dobney K, Brothwell D. 1986. Dental calculus: its relevance to ancient diet and oral ecology.
In: Cruwys E, Foley RA, editors. Teeth and anthropology. Oxford: BAR International
Series 291. p 55–81.

Dobney K, Brothwell D. 1988. A scanning electron microscope study of archaeological dental

calculus. In: Olsen SL, editor. Scanning electron microscopy in archaeology. Oxford:
BAR International Series. p 372–385.

Douglass GD, DeVreugd RT. 1997. The dynamics of occlusal relationships. In: McNeill C,
editor. Science and practice of occlusion. Berlin: Quintessence. p 69–78.

Dreher ML. 2012. Pistachio nuts: composition and potential health benefits. Nutr Rev
70:234–240.

Dubreuil L, Nadel D. 2015. The development of plant food processing in the Levant: insights
from use-wear analysis of Early Epipalaeolithic ground stone tools. Philos Trans R Soc
London B Biol Sci 370:20140357-.

Dudgeon J V, Tromp M. 2014. Diet, Geography and Drinking Water in Polynesia:

Microfossil Research from Archaeological Human Dental Calculus, Rapa Nui (Easter
Island). Int J Osteoarchaeol 24:634–648.

Dupont E. 1872. Les temps préhistoriques en Belgique: L’homme pendant les âges de la
pierre dans les environs de Dinant-sur-Meuse. Brussels: Murquardt.

Dusseldorp GL. 2010. Studying Pleistocene Neanderthal and cave hyena Dietary Habits:
Combining Isotopic and Archaeozoological Analyses. J Archaeol Method Theory
18:224–255.

139
Dusseldorp GL. 2013. Neanderthals and Cave Hyenas: Co-existence, Competition or
Conflict? In: Clark JL, Speth JD, editors. Zooarchaeology and Modern Human Origins.
Springer. p 191–208.

Ecker M, Bocherens H, Julien M-A, Rivals F, Raynal J-P, Moncel M-H. 2013. Middle
Pleistocene ecology and Neanderthal subsistence: insights from stable isotope analyses
in Payre (Ardèche, southeastern France). J Hum Evol 65:363–73.

Eliasson AC, Larsson K. 1993. Cereals in breadmaking: a molecular colloidal approach. New
York: CRC Press.

Evers AD. 1971. Scanning electron microscopy of wheat starch. III. Granule development in
the endosperm. Starch - Stärke 5:157–192.

Evins MA. 1982. The fauna from Shanidar Cave: Mousterian wild goat exploitation in
northeastern Iraq. Paléorient 8:37–58.

Fahy GE, Richards MP, Fuller BT, Deschner T, Hublin J-J, Boesch C. 2014. Stable nitrogen
isotope analysis of dentine serial sections elucidate sex differences in weaning patterns
of wild chimpanzees (Pan troglodytes). Am J Phys Anthropol 153:635–642.

Fahy GE, Richards MP, Riedel J, Hublin J-J, Boesch C. 2013. Stable isotope evidence of meat
eating and hunting specialization in adult male chimpanzees. Proc Natl Acad Sci U S A
110:5829–33.

Falchi M, El-Sayed Moustafa JS, Takousis P, Pesce F, Bonnefond A, Andersson-Assarsson JC,

Sudmant PH, Dorajoo R, Al-Shafai MN, Bottolo L, Ozdemir E, So H-C, Davies RW,
Patrice A, Dent R, Mangino M, Hysi PG, Dechaume A, Huyvaert M, Skinner J, Pigeyre
M, Caiazzo R, Raverdy V, Vaillant E, Field S, Balkau B, Marre M, Visvikis-Siest S, Weill
J, Poulain-Godefroy O, Jacobson P, Sjostrom L, Hammond CJ, Deloukas P, Sham PC,
McPherson R, Lee J, Tai ES, Sladek R, Carlsson LMS, Walley A, Eichler EE, Pattou F,
Spector TD, Froguel P. 2014. Low copy number of the salivary amylase gene
predisposes to obesity. Nat Genet 46:492–7.

Fenwick RSH, Lentfer CJ, Weisler MI. 2011. Palm reading: A pilot study to discriminate
phytoliths of four Arecaceae (Palmae) taxa. J Archaeol Sci 38:2190–2199.

Fernandes R, Millard AR, Brabec M, Nadeau M-J, Grootes P. 2014. Food Reconstruction
Using Isotopic Transferred Signals (FRUITS): A Bayesian Model for Diet
Reconstruction. PLoS One 9:e87436.

Field A. 2005. Discovering Statistics Using SPSS. Sage Publications Ltd.

Filhol MH. 1889. Note sur une machoire humaine trouvée dans la caverne de Malarnaud
près de Montseron (Ariège). Bull la Société Philomath Paris 8th 1:69–82.

Finlayson C. 2008. On the importance of coastal areas in the survival of Neanderthal

populations during the Late Pleistocene. Quat Sci Rev 27:2246–2252.

Finlayson C, Pacheco FG, Rodríguez-Vidal J, Fa DA, Gutierrez López JM, Santiago Pérez A,
Finlayson G, Allue E, Baena Preysler J, Cáceres I, Carrión JS, Fernández Jalvo Y, Gleed-

140
 Owen CP, Jiménez-Espejo FJ, López P, López Sáez JA, Riquelme Cantal JA, Sánchez
Marco A, Guzman FG, Brown K, Fuentes N, Valarino CA, Villalpando A, Stringer CB,
Martinez Ruiz F, Sakamoto T. 2006. Late survival of Neanderthals at the southernmost
extreme of Europe. Nature 443:850–3.

Fiore I, Gala M, Tagliacozzo A. 2004. Ecology and subsistence strategies in the eastern Italian
Alps during the Middle Palaeolithic. Int J Osteoarchaeol 14:273–286.

Fiorenza L. 2015. Reconstructing diet and behaviour of Neanderthals from Central Italy
through dental macrowear analysis. J Anthropol Sci 93:1–15.

Fiorenza L, Benazzi S, Henry AG, Salazar-García DC, Blasco R, Picin A, Wroe S, Kullmer O.
2015. To meat or not to meat? New perspectives on Neanderthal ecology. Am J Phys
Anthropol 156 Suppl:43–71.

Fiorenza L, Benazzi S, Tausch J, Kullmer O, Bromage TG, Schrenk F. 2011. Molar macrowear
reveals Neanderthal eco-geographic dietary variation. PLoS One 6:e14769.

Flannery K V. 1969. Origins and ecological effects of early domestication in Iran and the
Near East. In: Ucko PJ, Dimbleby GW, editors. The domestication and exploitation of
plants and animals. Chicago, IL: Gerald Duckworth and Co. p 73–100.

Flensborg G. 2011. Dento-alveolar lesions and palaeodietary inferences from the Paso Alsina
1 site (eastern Pampean-Patagonian transition, Argentina). Homo Int Zeitschrift für die
vergleichende Forsch am Menschen 62:335–50.

Floate MJS. 1970. Decomposition of organic materials from hill soils and pastures. Soil Biol
Biochem 2:173–185.

Foley R, Lahr MM. 2003. On stony ground: Lithic technology, human evolution, and the
emergence of culture. Evol Anthropol Issues, News, Rev 12:109–122.

Follieri M, Giardini M, Magri D, Sadori L. 1998. Palynostratigraphy of the Last Glacial

period in the volcanic region of central Italy. Quat Int 47–48:3–20.

Fox CL, Pérez-Pérez A. 1993. The diet of the Neanderthal Child Gibraltar 2 (Devil’s Tower)
through the study of the vestibular striation pattern. J Hum Evol 24:29–41.

Fox J, Weisberg S. 2002. An R Companion to Applied Regression. SAGE Publications.

Frayer DW, Fiore I, Lalueza-Fox C, Radovčić J, Bondioli L. 2010. Right handed Neandertals:
Vindija and beyond. J Anthropol Sci 88:113–127.

Fu Q, Hajdinjak M, Moldovan OT, Constantin S, Mallick S, Skoglund P, Patterson N,

Rohland N, Lazaridis I, Nickel B, Viola B, Prüfer K, Meyer M, Kelso J, Reich D, Pääbo S.
2015. An early modern human from Romania with a recent Neanderthal ancestor.
Nature 524:216–219.

Galván Santos B, Hernández CM, Ortega F. 2006. Territorio y producción lítica en los valles
de Alcoy (Alicante) durante el Paleolítico Medio. In: Martínez G, Morgado A, Alfonso
JA, editors. Sociedades prehistóricas, recursos abióticos y territorio. Granada:

141
 Fundación Ibn al-Jatib de Estudios de Cooperación Cultural. p 135–158.

Garralda MD, Galván B, Hernández CM, Mallol C, Gómez JA, Maureille B. 2014.
Neanderthals from El Salt (Alcoy, Spain) in the context of the latest Middle Palaeolithic
populations from the southeast of the Iberian Peninsula. J Hum Evol 75:1–15.

Gaudzinski-Windheuser S, Roebroeks W. 2011. On Neanderthal Subsistence in Last

Interglacial Forested Environments in Northern Europe. In: Conard NJ, Richter J,
editors. Neanderthal lifeways, subsistence and technology: one hundred fifty years of
Neanderthal study. Dordrecht: Springer Science & Business Media. p 61–71.

Germonpré M, Udrescu M, Fiers E. 2014. Possible evidence of mammoth hunting at the

Neanderthal site of Spy (Belgium). Quat Int 337:28–42.

Gobetz KE, Bozarth SR. 2001. Implications for Late Pleistocene mastodon diet from opal
phytoliths in tooth calculus. Quat Res 55:115–122.

Gomez-Robles A, Bermúdez de Castro JM, Arsuaga JL, Carbonell E, Polly PD. 2013. No
known hominin species matches the expected dental morphology of the last common
ancestor of Neanderthals and modern humans. Proc Natl Acad Sci U S A 110:18196–
18201.

Goodale N, Otis H, Andrefsky W, Kuijt I, Finlayson B, Bart K. 2010. Sickle blade life-history
and the transition to agriculture: an early Neolithic case study from Southwest Asia. J
Archaeol Sci 37:1192–1201.

Goodman MJ, Griffin PB, Estioko-Griffin AA, Grove JS. 1985. The compatibility of hunting
and mothering among the agta hunter-gatherers of the Philippines. Sex Roles 12:1199–
1209.

Granger J-M, Lévêque F. 1997. Parure castelperronienne et aurignacienne: étude de trois

séries inédites de dents percées et comparaisons. Comptes Rendus l’Académie des Sci -
Ser IIA - Earth Planet Sci 325:537–543.

Grayson DK, Delpech F. 2008. The large mammals of Roc de Combe (Lot, France): The
Châtelperronian and Aurignacian assemblages. J Anthropol Archaeol 27:338–362.

Greene TR, Kuba CL, Irish JD. 2005. Quantifying calculus: A suggested new approach for
recording an important indicator of diet and dental health. HOMO - J Comp Hum Biol
56:119–132.

Griggo C. 2004. Mousterian fauna from Dederiyeh Cave and comparisons with fauna from
Umm el Tlel and Douara Cave. Paléorient 30:149–162.

Grine FE. 1986. Dental evidence for dietary differences in Australopithecus and Paranthropus:
a quantitative analysis of permanent molar microwear. J Hum Evol 15:783–822.

Grün R, Stringer CB. 1991. Electron Spin Resonance Dating and the Evolution of Modern
Humans. Archaeometry 33:153–199.

Guatelli-Steinberg D, Larsen CS, Hutchinson DL. 2004. Prevalence and the duration of linear

142
 enamel hypoplasia: a comparative study of Neandertals and Inuit foragers. J Hum Evol
47:65–84.

Guérin G, Frouin M, Talamo S, Aldeias V, Bruxelles L, Chiotti L, Dibble HL, Goldberg P,

Hublin J-J, Jain M, Lahaye C, Madelaine S, Maureille B, McPherron SJP, Mercier N,
Murray AS, Sandgathe D, Steele TE, Thomsen KJ, Turq A. 2015. A multi-method
luminescence dating of the Palaeolithic sequence of La Ferrassie based on new
excavations adjacent to the La Ferrassie 1 and 2 skeletons. J Archaeol Sci 58:147–166.

Gunz P, Neubauer S, Maureille B, Hublin J-J. 2010. Brain development after birth differs
between Neanderthals and modern humans. Curr Biol 20:R921–R922.

Gurven M, Allen-Arave W, Hill K, Hurtado AM. 2001. Reservation food sharing among the
Ache of Paraguay. Hum Nat 12:273–297.

Guthrie DR. 2001. Origin and causes of the mammoth steppe: a story of cloud cover, woolly
mammal tooth pits, buckles, and inside-out Beringia. Quat Sci Rev 20:549–574.

Haber Uriarte M, Avilés Fernández A, Lomba Maurandi J. 2013. Estudio antropológico

preliminar de los restos humanos calcolíticos del enterramiento múltiple de Camino del
Molino (Caravaca de la Cruz, Murcia). In: Turbón D, Fañanás L, Rissech C, Rosa A,
editors. Biodiversidad humana y evolución. Barcelona: Universidad de Barcelona. p
236–242.

Halstead P. 2012. Feast, Food and Fodder in Neolithic-Bronze Age Greece: Commensality
and the Construction of Value. eTopoi J Anc Stud 2:21–51.

Hardy BL. 2010. Climatic variability and plant food distribution in Pleistocene Europe:
Implications for Neanderthal diet and subsistence. Quat Sci Rev 29:662–679.

Hardy BL, Kay M, Marks AE, Monigal K. 2001. Stone tool function at the paleolithic sites of
Starosele and Buran Kaya III, Crimea: Behavioral implications. Proc Natl Acad Sci U S
A 98:10972–10977.

Hardy BL, Moncel M-H. 2011. Neanderthal use of fish, mammals, birds, starchy plants and
wood 125-250,000 years ago. PLoS One 6:e23768.

Hardy BL, Moncel M-H, Daujeard C, Fernandes P, Béarez P, Desclaux E, Chacon Navarro
MG, Puaud S, Gallotti R. 2013. Impossible Neanderthals? Making string, throwing
projectiles and catching small game during Marine Isotope Stage 4 (Abri du Maras,
France). Quat Sci Rev 82:23–40.

Hardy K, Blakeney T, Copeland L, Kirkham J, Wrangham RW, Collins MJ. 2009. Starch
granules, dental calculus and new perspectives on ancient diet. J Archaeol Sci 36:248–
255.

Hardy K, Brand-Miller J, Brown KD, Thomas MG, Copeland L. 2015a. The Importance of
Dietary Carbohydrate in Human Evolution. Q Rev Biol 90:251–268.

Hardy K, Buckley S, Collins MJ, Estalrrich A, Brothwell D, Copeland L, García-Tabernero A,

García-Vargas S, de la Rasilla M, Lalueza-Fox C, Huguet R, Bastir M, Santamaría D,

143
 Madella M, Wilson J, Cortés AF, Rosas A. 2012. Neanderthal medics? Evidence for
food, cooking, and medicinal plants entrapped in dental calculus. Naturwissenschaften
99:617–26.

Hardy K, Radini A, Buckley S, Sarig R, Copeland L, Gopher A, Barkai R. 2015b. Dental

calculus reveals potential respiratory irritants and ingestion of essential plant-based
nutrients at Lower Palaeolithic Qesem Cave Israel. Quat Int.

Harlan JR. 1989. Wild-grass seed harvesting in the Sahara and sub-Sahara of Africa. In:
Harris DR, Hillman GC, editors. Foraging and farming: the evolution of plant
exploitation. London: Unwin Hyman. p 79–98.

Harvati K, Darlas A, Bailey SE, Rein TR, El Zaatari S, Fiorenza L, Kullmer O, Psathi E. 2013.
New Neanderthal remains from Mani peninsula, Southern Greece: the Kalamakia
Middle Paleolithic cave site. J Hum Evol 64:486–99.

Harvati K, Panagopoulou E, Runnels C. 2009. The Paleoanthropology of Greece. Evol

Anthropol 18:131–143.

Haslam M. 2004. The decomposition of starch grains in soils: implications for archaeological
residue analyses. J Archaeol Sci 31:1715–1734.

Hawkes K, Bird RB. 2002. Showing Off, Handicap Signaling, and the Evolution of Men’s
Work. Evol Anthropol 11:58–67.

Hawkes K, Hill K, O’Connell JF. 1982. Why hunters gather: optimal foraging and the Ache of
eastern Paraguay. Am Ethnol 9:379–398.

Haws JA. 2004. An Iberian perspective on Upper Paleolithic plant consumption.

Promontoria 2:49–105.

Haws JA, Hockett BS. 2004. Theoretical perspectives on the dietary role of small animals in
human evolution. Petits Animaux Soc Hum:173–184.

Hedges REM, Clement JG, Thomas CDL, O’Connell TC. 2007. Collagen turnover in the adult
femoral mid-shaft: modeled from anthropogenic radiocarbon tracer measurements. Am
J Phys Anthropol 133:808–16.

Henri-Martin G. 1961. Le niveau de Chatelperron a la Quina (Charente). Bull la Société

Préhistorique Française 58:796–808.

Henry AG. 2010. Plant Foods and The Dietary Ecology of Neandertals and Modern Humans.

Henry AG. 2014. Formation and taphonomic processes affecting starch grains. In: Marston
JM, Guedes JD, Warinner C, editors. Current Methods in Paleoethnobotany. Boulder:
University of Colorado Press. p 35–50.

Henry AG, Brooks AS, Piperno DR. 2011. Microfossils in calculus demonstrate consumption
of plants and cooked foods in Neanderthal diets (Shanidar III, Iraq; Spy I and II,
Belgium). Proc Natl Acad Sci U S A 108:486–491.

144
Henry AG, Brooks AS, Piperno DR. 2014. Plant foods and the dietary ecology of
Neanderthals and early modern humans. J Hum Evol 69:44–54.

Henry AG, Piperno DR. 2008. Using plant microfossils from dental calculus to recover
human diet: a case study from Tell al-Raqā’i, Syria. J Archaeol Sci 35:1943–1950.

Henry AG, Ungar PS, Passey BH, Sponheimer M, Rossouw L, Bamford M, Sandberg P, de
Ruiter DJ, Berger L. 2012. The diet of Australopithecus sediba. Nature 487:90–3.

Heyes P, MacDonald K. 2015. Neandertal energetics: Uncertainty in body mass estimation

limits comparisons with Homo sapiens. J Hum Evol 85:193–7.

Higham TFG. 2011. European Middle and Upper Palaeolithic radiocarbon dates are often
older than they look: problems with previous dates and some remedies. Antiquity
85:235–249.

Higham TFG, Douka K, Wood RE, Ramsey CB, Brock F, Basell L, Camps M, Arrizabalaga A,
Baena J, Barroso-Ruíz C, Bergman C, Boitard C, Boscato P, Caparrós M, Conard NJ,
Draily C, Froment A, Galván B, Gambassini P, Garcia-Moreno A, Grimaldi S, Haesaerts
P, Holt B, Iriarte-Chiapusso M-J, Jelinek AJ, Jordá Pardo JF, Maíllo-Fernández J-M,
Marom A, Maroto J, Menéndez M, Metz L, Morin E, Moroni A, Negrino F,
Panagopoulou E, Peresani M, Pirson S, de la Rasilla M, Riel-Salvatore J, Ronchitelli A,
Santamaría D, Semal P, Slimak L, Soler J, Soler N, Villaluenga A, Pinhasi R, Jacobi R.
2014. The timing and spatiotemporal patterning of Neanderthal disappearance. Nature
512:306–9.

Higham TFG, Jacobi R, Julien M, David F, Basell L, Wood RE, Davies W, Ramsey CB. 2010.
Chronology of the Grotte du Renne (France) and implications for the context of
ornaments and human remains within the Châtelperronian. Proc Natl Acad Sci U S A
107:20234–9.

Higham TFG, Ramsey CB, Karavanić I, Smith FH, Trinkaus E. 2006. Revised direct
radiocarbon dating of the Vindija G1 Upper Paleolithic Neandertals. Proc Natl Acad Sci
U S A 103:553–7.

Hill K. 1988. Macronutrient modifications of optimal foraging theory: An approach using

indifference curves applied to some modern foragers. Hum Ecol 16:157–197.

Hockett BS. 2012. The consequences of Middle Paleolithic diets on pregnant Neanderthal
women. Quat Int 264:78–82.

Hockett BS, Haws JA. 2003. Nutritional Ecology and Diachronic Trends in Paleolithic Diet
and Health. Evol Anthropol 12:211–216.

Hockett BS, Haws JA. 2005. Nutritional ecology and the human demography of Neandertal
extinction. Quat Int 137:21–34.

Hockett BS, Haws JA. 2009. Continuity in animal resource diversity in the Late Pleistocene
human diet of Central Portugal. Before Farming 2009:1–14.

Hohmann G, Potts K, N’Guessan A, Fowler A, Mundry R, Ganzhorn JU, Ortmann S. 2010.

145
 Plant foods consumed by Pan: exploring the variation of nutritional ecology across
Africa. Am J Phys Anthropol 141:476–85.

Holm G. 1911. Ethnological sketch of the Angmagssalik Eskimo. In: Thalbitzer W, editor.
The Ammassalik Eskimo : contributions to the Ethnology of the East Greenland
Natives. Copenhagen: Meddelelser om Grønland. p 1–147.

Holm J, Lundquist I, Björck I, Eliasson AC, Asp NG. 1988. Degree of starch gelatinization,
digestion rate of starch in vitro, and metabolic response in rats. Am J Clin Nutr 47:1010–
6.

Hora M, Sladek V. 2014. Influence of lower limb configuration on walking cost in Late
Pleistocene humans. J Hum Evol 67:19–32.

Howell FC. 1957. The Evolutionary Significance of Variation and Varieties of “Neanderthal”
Man. Q Rev Biol 32:330–347.

Hublin J-J. 1998. Climatic changes, paleogeography, and the evolution of the Neandertals.
In: Akazawa T, Aoki K, Bar-Yosef O, editors. Neandertals and modern humans in
Western Asia. New York, NY: Kluwer Academic Publishers. p 295–310.

Hublin J-J. 2009. The origin of Neandertals. Proc Natl Acad Sci U S A 106:16022–7.

Hublin J-J. 2015. The modern human colonization of western Eurasia: when and where?
Quat Sci Rev 118:194–210.

Hublin J-J, Roebroeks W. 2009. Ebb and flow or regional extinctions? On the character of
Neandertal occupation of northern environments. Comptes Rendus Palevol 8:503–509.

Hublin J-J, Spoor F, Braun M, Zonneveld F, Condemi S. 1996. A late Neanderthal associated
with Upper Palaeolithic artefacts. Nature 381:224–6.

Hublin J-J, Talamo S, Julien M, David F, Connet N, Bodu P, Vandermeersch B, Richards MP.
2012. Radiocarbon dates from the Grotte du Renne and Saint-Césaire support a
Neandertal origin for the Châtelperronian. Proc Natl Acad Sci U S A 109:18743–8.

Hublin J-J, Weston D, Gunz P, Richards M, Roebroeks W, Glimmerveen J, Anthonis L. 2009.

Out of the North Sea: the Zeeland ridges Neandertal. J Hum Evol 57:777–85.

Hughes DR. 1963. A study of a series of mandibles from the Mulu cave, Sarawak. J
Anthropol Inst Gt Britain Irel 93:235–249.

Hurtado MA, Hill KR. 1987. Early dry season subsistence ecology of Cuiva (Hiwi) foragers
of Venezuela. Hum Ecol 15:163–187.

Hutchinson GE. 1959. Homage to Santa Rosalia or why are there so many kinds of animals?
Am Nat 93:145–159.

ICSN. 2011. The International Code for Starch Nomenclature. Available from:
http://fossilfarm.org/ICSN/Code.html

Janis CM. 1990. The correlation between diet and dental wear in herbivorous mammals, and

146
 its relationship to the determination of diets of extinct species. In: Boucot AJ, editor.
Evolutionary paleobiology of behavior and coevolution. Amsterdam: Elsevier. p 241–
259.

Jin Y, Yip H-K. 2002. Supragingival calculus: formation and control. Crit Rev Oral Biol Med
13:426–441.

Jones EL. 2016. In Search of the Broad Spectrum Revolution in Paleolithic Southwest Europe.
Cham: Springer International Publishing.

Jones M. 2009. Moving North: Archaeobotanical Evidence for Plant Diet in Middle and
Upper Paleolithic Europe. In: Hublin J-J, Richards MP, editors. The Evolution of
Human Diets: Integrating Approaches to the Study of Paleolithic Subsistence. Springer.
p 171–180.

Kaidonis JA. 2008. Tooth wear: the view of the anthropologist. Clin Oral Investig 12 Suppl
1:S21-6.

Kaifu Y, Kasai K, Townsend GC, Richards LC. 2003. Tooth Wear and the “Design” of the
Human Dentition: A Perspective from Evolutionary Medicine. Am J Phys Anthropol
122:47–61.

Kaplan AR, Powell WE, Moorhouse AB, Hinko EN. 1976. Taste sensitivity and human
variation: some biological and clinical implications. In: Kaplan AR, editor. Human
Behaviour Genetics. Springfield: Charles C. Thomas. p 401–423.

Karavanić I, Smith FH. 1998. The Middle/Upper Paleolithic interface and the relationship of
Neanderthals and early modern humans in the Hrvatsko Zagorje, Croatia. J Hum Evol
34:223–248.

Kay RF, Hiiemae KM. 1974. Jaw movement and tooth use in recent and fossil primates. Am J
Phys Anthropol 40:227–56.

Keeley LH. 1992. The Use of Plant Foods among Hunter-Gatherers : A Cross-Cultural
Survey. In: Anderson PC, editor. Prehistoire de l’Agriculture. Nouvelles Approches
Experimentales et Ethnographiques. Paris: National Center for Scientific Research.
Monographie du CRA. p 29–38.

Kelly RL. 1995. The Foraging Spectrum: Diversity in Hunter-Gatherer Lifeways. New York:
Eliot Werner Publications.

Klein RG. 2009. The human career: human biological and cultural origins. Chicago:
University of Chicago Press.

Klein RG, Steele TE. 2008. Gibraltar data are too sparse to inform on Neanderthal
exploitation of coastal resources. Proc Natl Acad Sci U S A 105:E115; author reply E116.

Kleinberg I. 1970. Biochemistry of the Dental Plaque. In: Staple PH, editor. Advances in Oral
Biology, Volume 4. New York: Elsevier Science. p 40–93.

Krause J, Orlando L, Serre D, Viola B, Prüfer K, Richards MP, Hublin J-J, Hänni C,

147
 Derevianko AP, Pääbo S. 2007. Neanderthals in central Asia and Siberia. Nature
449:902–4.

Krings M, Capelli C, Tschentscher F, Geisert H, Meyer S, von Haeseler A, Grossschmidt K,

Possnert G, Paunović M, Pääbo S. 2000. A view of Neandertal genetic diversity. Nat
Genet 26:144–146.

Kucera M, Pany-Kucera D, Boyadjian CHC, Reinhard K, Eggers S. 2011. Efficient but

destructive: A test of the dental wash technique using secondary electron microscopy. J
Archaeol Sci 38:129–135.

Kuhn SL. 1991. “Unpacking” reduction: Lithic raw material economy in the mousterian of
west-central Italy. J Anthropol Archaeol 10:76–106.

Kuhn SL, Stiner MC. 2006. What’s a Mother to Do? The Division of Labor among
Neandertals and Modern Humans in Eurasia. Curr Anthropol 47:953–981.

Kuhnlein H V., Turner NJ. 1991. Traditional plant foods of Canadian indigenous peoples:
nutrition, botany, and use v8. Amsterdam: Taylor & Francis.

Kuitems M, van Kolfschoten T, van der Plicht J. 2012. Elevated δ15N values in mammoths: a
comparison with modern elephants. Archaeol Anthropol Sci 7:289–295.

Kullmer O, Benazzi S, Fiorenza L, Schulz D, Bacso S, Winzen O. 2009. Technical note:

Occlusal fingerprint analysis: Quantification of tooth wear pattern. Am J Phys
Anthropol 139:600–605.

Laden G, Wrangham RW. 2005. The rise of the hominids as an adaptive shift in fallback
foods: plant underground storage organs (USOs) and australopith origins. J Hum Evol
49:482–498.

Lalueza C, Pérez-Pérez A, Turbón D. 1996. Dietary inferences through buccal microwear

analysis of middle and upper Pleistocene human fossils. Am J Phys Anthropol 100:367–
87.

Lalueza-Fox C, Gigli E, de la Rasilla M, Fortea J, Rosas A. 2009. Bitter taste perception in

Neanderthals through the analysis of the TAS2R38 gene. Biol Lett 5:809–11.

Lalueza-Fox C, Juan J, Albert RM. 1996. Phytolith analysis on dental calculus, enamel
surface, and burial soil: Information about diet and paleoenvironment. Am J Phys
Anthropol 101:101–113.

Lanfranco LP, Eggers S. 2012. Caries through time: an anthropological overview. In: Li M-Y,
editor. Contemporary Approach to Dental Caries. InTech. p 3–34.

Langejans GHJ. 2010. Remains of the day-preservation of organic micro-residues on stone

tools. J Archaeol Sci 37:971–985.

Langejans GHJ. 2011. Discerning use-related micro-residues on tools: testing the multi-
stranded approach for archaeological studies. J Archaeol Sci 38:985–1000.

148
Langejans GHJ. 2012. Middle Stone Age pièces esquillées from Sibudu Cave, South Africa:
an initial micro-residue study. J Archaeol Sci 39:1694–1704.

Larsen CS, Shavit R, Griffin MC. 1991. Dental caries evidence for dietary change: an
archaeological context. In: Kelley M, Larsen C, editors. Advances in Dental
Anthropology. New York, NY: Wiley-Liss. p 179–202.

Laurence AR, Thoms A V., Bryant VM, McDonough C. 2011. Airborne Starch Granules as a
Potential Contamination Source at Archaeological Sites. J Ethnobiol 31:213–232.

Lazzati AMB, Levrini L, Rampazzi L, Dossi C, Castelletti L, Licata M, Corti C. 2015. The Diet
of Three Medieval Individuals from Caravate (Varese, Italy). Combined Results of ICP-
MS Analysis of Trace Elements and Phytolith Analysis Conducted on Their Dental
Calculus. Int J Osteoarchaeol.

Lebreton V, Psathi E, Darlas A. 2008. Environnement vegetal des neandertaliens de la Grotte

deKalamakia (Aréopolis, Grèce). In: Darlas A, Mihailovi D, editors. The Palaeolithic of
the Balkan. Proceedings of the XV UISPP World Congress. Oxford: BAR International
Series 1819. p 61–68.

Lee B-H, Bello-Pérez LA, Lin AH-M, Kim CY, Hamaker BR. 2013. Importance of Location of
Digestion and Colonic Fermentation of Starch Related to Its Quality. Cereal Chem
90:335–343.

Lee-Thorp JA, van der Merwe NJ, Brain CK. 1994. Diet of Australopithecus robustus at
Swartkrans from stable carbon isotopic analysis. J Hum Evol 27:361–372.

Lee-Thorp J, Sponheimer M. 2006. Contributions of biogeochemistry to understanding

hominin dietary ecology. Yearb Phys Anthropol 43:131–148.

Leigh RW. 1925. Dental pathology of Indian tribes of varied environmental and food
conditions. Am J Phys Anthropol 8:179–199.

Leonard C, Vashro L, O’Connell JF, Henry AG. 2015. Plant microremains in dental calculus
as a record of plant consumption: A test with Twe forager-horticulturalists. J Archaeol
Sci Reports 2:449–457.

Leroi-Gourhan A. 1968. Le Neanderthalien IV de Shanidar. Bull la Société Préhistorique

Française LXV:79–83.

Leroi-Gourhan A. 1969. Pollen grains of Graminae and Cerealia from Shanidar and Zawi
Chemi. In: Ucko PJ, Dimbleby GW, editors. The domestication and exploitation of
plants and animals. Chicago: Aldine Transaction. p 141–148.

Leroi-Gourhan A. 1975. The Flowers Found with Shanidar IV, a Neanderthal Burial in Iraq.
Science (80- ) 190:562–564.

Lev E, Kislev ME, Bar-Yosef O. 2005. Mousterian vegetal food in Kebara Cave, Mt. Carmel. J
Archaeol Sci 32:475–484.

Lieverse AR. 1999. Diet and the Aetiology of Dental Calculus. Int J Osteoarchaeol 232:219–

149
 232.

Liu H. 2012. Research of plant remains of some sites in Shaanxi and Henan and their
implication for the two millets of China.

Liu L, Bestel S, Shi J, Song Y, Chen X. 2013. Paleolithic human exploitation of plant foods
during the last glacial maximum in North China. Proc Natl Acad Sci U S A 110:5380–
5385.

Lomba Maurandi J, López Martínez M, Ramos Martínez F, Avilés Fernández A. 2009. El

enterramiento múltiple, calcolítico, de Camino del Molino (Caravaca, Murcia).
Metodología y primeros resultados de un yacimiento exepcional. Trab Prehist 66:143–
159.

Loring Brace C, Rosenberg KR, Hunt KD. 1987. Gradual change in human tooth size in the
late Pleistocene and post-Pleistocene. Evolution (N Y) 41:705–720.

Lothrop SK. 1928. The Indians of Tierra del Fuego. New York: Museum of the American
Indian.

Loy T, Barton H. 2006. Post-excavation contamination and measures for prevention. In:
Torrence R, Barton H, editors. Ancient Starch Research. Walnut Creek: Left Coast Press.
p 165.

Lucas PW, Omar R, Al-Fadhalah K, Almusallam AS, Henry AG, Michael S, Thai LA, Watzke
J, Strait DS, Atkins AG. 2013. Mechanisms and causes of wear in tooth enamel:
implications for hominin diets. J R Soc Interface 10:20120923.

Lukacs JR, Largaespada LL. 2006. Explaining sex differences in dental caries prevalence:
saliva, hormones, and “life-history” etiologies. Am J Hum Biol 18:540–55.

De Lumley H, Darlas A, Anglada R, Cataliotti-Valdina J, Desclaux E, Dubar M, Falguères C,

Keraudren B, Lecervoisier B, Mestour B, Renault-Miskovsky J, Trantalidou K, Vernet
JL. 1994. Grotte de Kalamakia (Aréopolis, Péloponnèse). Bull Corresp héllénique
118:535–559.

Lyman RL. 2003. The influence of time averaging and space averaging on the application of
foraging theory in zooarchaeology. J Archaeol Sci 30:595–610.

MacArthur RH, Pianka ER. 1966. On Optimal Use of a Patchy Environment. Am Nat
100:603.

MacDonald IA, Bokkenheuser VD, Winter J, McLernon AM, Mosbach EH. 1983.
Degradation of steroids in the human gut. J Lipid Res 24:675–700.

Macdonald K, Roebroeks W, Verpoorte A. 2009. An Energetics Perspective on the

Neandertal Record. In: Hublin J-J, Richards MP, editors. The Evolution of Hominin
Diets: Integrating Approaches to the Study of Palaeolithic Subsistence. Dordrecht:
Springer. p 211–220.

Madella M, Alexandre A, Ball T. 2005. International code for phytolith nomenclature 1.0.

150
 Ann Bot 96:253–260.

Madella M, Jones MK, Goldberg P, Goren Y, Hovers E. 2002. The Exploitation of Plant
Resources by Neanderthals in Amud Cave (Israel): The Evidence from Phytolith
Studies. J Archaeol Sci 29:703–719.

Magurran AE. 2004. Measuring biological diversity. Oxford: Blackwell Publishing.

Mangafa M. 1998. Plant exploitation from the Middle Paleolithic to the Neolithic: from food
gathering to farming. Archaeobotanical study of Theopetra cave. Theopetra Cave Proc
Int Conf Trkala:6–7.

Marlowe FW. 2010. The Hadza: hunter-gatherers of Tanzania. London: University of

California Press.

Martinoli DC. 2005. Plant food economy and environment during the epipalaeolithic in
southwest Anatolia : an investigation of the botanical macroremains from Öküzini and
Karain B.

Martinson DG, Pisias NG, Hays JD, Imbrie J, Moore TC, Shackleton NJ. 1987. Age dating
and the orbital theory of the ice ages: Development of a high-resolution 0 to 300,000-
year chronostratigraphy. Quat Res 27:1–29.

Matsutani A. 1987. Plant remains from the 1984 excavations at Douara Cave. In: Akazawa T,
Sakaguchi Y, editors. Paleolithic Site of Douara Cave and Paleogeography of Palmyra
Basin in Syria: Part IV: 1984 Excavations. Tokyo: The University of Tokyo Press.

Mauch Lenardić J. 2014. Bank vole Myodes (=Clethrionomys) glareolus (Schreber, 1780):
Rare species in the Late Pleistocene fauna of Croatia. Quat Int 328–329:167–178.

McBrearty S, Brooks AS. 2000. The revolution that wasn’t: a new interpretation of the origin
of modern human behavior. J Hum Evol 39:453–563.

Mellars P. 1986. A new chronology for the French Mousterian period. Nature 322:410–411.

Mercader J, Bennett T, Raja M. 2008. Middle Stone Age starch acquisition in the Niassa Rift,
Mozambique. Quat Res 70:283–300.

Meriggi A, Rosa P, Brangi A, Matteucci C. 1991. Habitat use and diet of the wolf in northern
Italy. Acta Theriol (Warsz) 36:141–151.

Meyer M, Fu Q, Aximu-Petri A, Glocke I, Nickel B, Arsuaga J-L, Martinez I, Gracia A, de

Castro JMB, Carbonell E, Paabo S. 2014. A mitochondrial genome sequence of a
hominin from Sima de los Huesos. Nature 505:403–406.

Mickleburgh HL, Pagán-Jiménez JR. 2012. New insights into the consumption of maize and
other food plants in the pre-Columbian Caribbean from starch grains trapped in human
dental calculus. J Archaeol Sci 39:2468–2478.

Molleson TI. 2000. The People of Abu Hureyra. In: Moore AMT, Hillman CC, Legge AJ,
editors. Village on the Euphrates from Foraging to Farming at Abu Hureyra. Oxford:

151
 Oxford University Press. p 301–34.

Monge G, Jiménez-Espejo FJ, García-Alix A, Martínez-Ruiz F, Mattielli N, Finlayson C,

Ohkouchi N, Sánchez MC, de Castro JMB, Blasco R, Rosell J, Carrión JS, Rodríguez-
Vidal J, Finlayson G. 2015. Earliest evidence of pollution by heavy metals in
archaeological sites. Sci Rep 5:14252.

Monteiro da Silva A. M, Newman HN, Oakley DA, O’Leary R. 1998. Psychosocial factors,
dental plaque levels and smoking in periodontitis patients. J Clin Periodontol 25:517–
523.

Morton JF. 1975. Cattails (Typha spp.) — Weed Problem or Potential Crop? Econ Bot 29:7–29.

Mulder MB, Schacht R. 2012. Human Behavioural Ecology. eLS.

Murie A. 1944. The wolves of Mount McKinley. Fauna of the National Parks of the U. S.
Fauna Series No. 5. Fauna Series No. 5.

Murray J, Nasheuer HP, Seoighe C, McCormack GP, Williams DM, Harper DAT. 2015. The
Contribution of William King to the Early Development of Palaeoanthropology. Irish J
Earth Sci 33:1–16.

N’guessan AK. 2012. Aspects quantitatifs et qualitatifs du régime alimentaire des

chimpanzés au parc national de Taï, Côte d’Ivoire.

N’guessan AK, Ortmann S, Boesch C. 2009. Daily energy balance and protein gain among
Pan troglodytes verus in the Taï National Park, Côte d’Ivoire. Int J Primatol 30:481–496.

Nabhan GP. 2009. Ethnoecology: Bridging Disciplines, Cultures and Species. J Ethnobiol
29:3–7.

Nadel D, Danin A, Power RC, Rosen AM, Bocquentin F, Tsatskin A, Rosenberg D, Yeshurun
R, Weissbrod L, Rebollo NR, Barzilai O, Boaretto E. 2013. Earliest floral grave lining
from 13,700-11,700-y-old Natufian burials at Raqefet Cave, Mt. Carmel, Israel. Proc Natl
Acad Sci U S A 110:1–5.

Newman AW, Vitez IM, Kiesnowski C, Mueller RL. 1996. Starches and starch derivatives. In:
Swarbrick J, Boylan JC, editors. Encyclopedia of pharmaceutical technology. New York:
Marcel Dekker Inc. p 223–248.

Newsom SWB, Shaw M. 1997. A survey of starch particle counts in the hospital environment
in relation to the use of powdered latex gloves. Occup Med (Chic Ill) 47:155–158.

Nicholson RA. 1996. Bone Degradation, Burial Medium and Species Representation:
Debunking the Myths, an Experiment-based Approach. J Archaeol Sci 23:513–533.

Niewind A, Krondl M, Shrott M. 1988. Genetic influences on the selection of brassica

vegetables by elderly individuals. Nutr Res 8:13–20.

Nilles J. 1942. Digging-Sticks, Spades, Hoes, Axes, and Adzes of the Kuman People in the
Bismarck Mountains of East-Central New Guinea. Anthropos 37/40:205–212.

152
Nishida T, Hasegawa T. 1992. Meat-sharing as a coalition strategy by an alpha male
chimpanzee. In: Nishida T, Mcgrew WC, Marler P, Pickford M, de Waal MFB, editors.
Topics in primatology. Tokyo: University of Tokyo Press. p 159–174.

O’Connell JF. 2006. How did modern humans displace Neanderthals? Insights from hunter-
gatherer ethnography and archaeology. In: Conard NJ, editor. When Neanderthals and
modern humans met. Tübingen: Kerns Verlag. p 43–64.

O’Connell JF, Hawkes K, Blurton Jones NG. 1999. Grandmothering and the evolution of
homo erectus. J Hum Evol 36:461–85.

Odum EP. 1975. Ecology. 2nd Editio. New York: Holt, Rinehart and Winston.

Otte M. 1979. Le Paléolithique supérieur ancien en Belgique. Bruxelles: Musees Royaux d’art
et d’histoire.

Out WA, Madella M. 2015. Morphometric distinction between bilobate phytoliths from
Panicum miliaceum and Setaria italica leaves. Archaeol Anthropol Sci.

Out WA, Pertusa Grau JF, Madella M. 2014. A new method for morphometric analysis of
opal phytoliths from plants. Microsc Microanal 20:1876–87.

Ovchinnikov I V., Götherström A, Romanova GP, Kharitonov VM, Lidén K, Goodwin W.

2000. Molecular analysis of Neanderthal DNA from the northern Caucasus. Nature
404:490–3.

Oyebade T. 1973. Some aspects of developmental physiology of the Nigerian kola (Cola
nitida) fruit. Econ Bot 27:417–422.

Pacher M, Stuart AJ. 2009. Extinction chronology and palaeobiology of the cave bear (Ursus
spelaeus). Boreas 38:189–206.

Patou-Mathis M. 2000. Neanderthal subsistence behaviours in Europe. Int J Osteoarchaeol

10:379–395.

Paunović M, Smith FH. 2002. Taphonomy of lower vertebrates from Vindija cave (Croatia):
delicacy on Neandertal table of animal prey? J Hum Evol 42:A27–A27.

Pennington HL, Weber SA. 2004. Paleoethnobotany: Modern Research Connecting Ancient
Plants and Ancient Peoples. CRC Crit Rev Plant Sci 23:13–20.

Pennington L. 1989. Bowes and Church’s Food values of portions commonly used. New
York: Harper and Row.

Pérez Jordà G. 2005. Nuevos datos paleocarpológicos en niveles neolíticos del País
Valenciano. In: Arias Cabal P, Ontañón Peredo R, García-Moncó Piñeiro C, editors. III
Congreso del Neolítico en la Península Ibérica. Universidad de Cantabria. p 103–114.

Pérez Jordà G, Carrión Marco Y. 2011. Los recursos vegetales. In: Pérez Jordà G, Bernabeu
Aubán J, Carrión Marco Y, García Puchol O, Molina Balaguer LL, Gómez Puche M,
editors. La Vital (Gandia, Valencia). Vida y muerte en la desembocadura del Serpis

153
 durante el III y el I milenio a.C. Museo de Prehistòria de València-Diputación de
Valencia.

Pérez-Pérez A. 1994. Intraindividual and intragroup variability of buccal tooth striation

pattern. Am J Phys Anthropol 94:175–187.

Perkins D. 1964. Prehistoric Fauna From Shanidar, Iraq. Science (80- ) 144:1565–1566.

Perry GH, Dominy NJ, Claw KG, Lee AS, Fiegler H, Redon R, Werner J, Villanea FA,
Mountain JL, Misra R, Carter NP, Lee C, Stone AC. 2007. Diet and the evolution of
human amylase gene copy number variation. Nat Genet 39:1256–1260.

Perry GH, Kistler L, Kelaita MA, Sams AJ. 2015. Insights into hominin phenotypic and
dietary evolution from ancient DNA sequence data. J Hum Evol 79:55–63.

Peters KE, Walters CC, Moldowan JM. 2005. The Biomarker Guide. Volume 1: Biomarkers
and Isotopes in the Environment and Human History. Cambridge University Press.

Phillips C, Lancelotti C. 2014. Chimpanzee diet: phytolithic analysis of feces. Am J Primatol

76:757–73.

Phinney SD. 1995. The functional effects of carbohydrate and energy underconsumption. In:
Bernadette MM, editor. Not Eating Enough: Overcoming Underconsumption of
Military Operational Rations. Washington, DC: National Academy Press. p 303–315.

Pinhasi R, Higham TFG, Golovanova L V., Doronichev VB. 2011. Revised age of late
Neanderthal occupation and the end of the Middle Paleolithic in the northern
Caucasus. Proc Natl Acad Sci U S A 108:8611–6.

Piperno DR. 2006. Phytoliths: A Comprehensive Guide for Archaeologists and

Paleoecologists. Lanham: AltaMira.

Piperno DR, Dillehay TD. 2008. Starch grains on human teeth reveal early broad crop diet in
northern Peru. Proc Natl Acad Sci U S A 105:19622–7.

Power RC, Rosen AM, Nadel D. 2014a. The economic and ritual utilization of plants at the
Raqefet Cave Natufian site: The evidence from phytoliths. J Anthropol Archaeol 33:49–
65.

Power RC, Salazar-García DC, Henry, Amanda G. 2016. Dental calculus evidence of
Gravettian diet and behaviour at Dolní Věstonice and Pavlov. In: Svoboda J, editor.
Dolní Věstonice II: Chronostratigraphy, Paleoethnology, Paleoanthropology. Brno:
Academy of Sciences of the Czech Republic, Institute of Archeology,. p 345–352.

Power RC, Salazar-García DC, Straus LG, González Morales MR, Henry AG. 2015a.
Microremains from El Mirón on Cave human dental calculus suggest a mixed plant-
animal subsistence economy during the Magdalenian in Northern Iberia. J Archaeol Sci
60:39–46.

Power RC, Salazar-García DC, Wittig RM, Freiberg M, Henry AG. 2015b. Dental calculus
evidence of Taï Forest Chimpanzee plant consumption and life history transitions. Sci

154
 Rep 5:15161.

Power RC, Salazar-García DC, Wittig RM, Henry AG. 2014b. Assessing use and suitability of
scanning electron microscopy in the analysis of micro remains in dental calculus. J
Archaeol Sci 49:160–169.

Privat KL, O’Connell TC, Hedges REM. 2007. The distinction between freshwater- and
terrestrial-based diets: methodological concerns and archaeological applications of
sulphur stable isotope analysis. J Archaeol Sci 34:1197–1204.

Prüfer K, Racimo F, Patterson N, Jay F, Sankararaman S, Sawyer S, Heinze A, Renaud G,

Sudmant PH, de Filippo C, Li H, Mallick S, Dannemann M, Fu Q, Kircher M, Kuhlwilm
M, Lachmann M, Meyer M, Ongyerth M, Siebauer M, Theunert C, Tandon A, Moorjani
P, Pickrell J, Mullikin JC, Vohr SH, Green RE, Hellmann I, Johnson PLF, Blanche H,
Cann H, Kitzman JO, Shendure J, Eichler EE, Lein ES, Bakken TE, Golovanova L V.,
Doronichev VB, Shunkov M V., Derevianko AP, Viola B, Slatkin M, Reich D, Kelso J,
Pääbo S. 2014. The complete genome sequence of a Neanderthal from the Altai
Mountains. Nature 505:43–9.

Pryor AJE, Steele M, Jones MK, Svoboda J, Beresford-Jones DG. 2013. Plant foods in the
Upper Palaeolithic at Dolní Vӗstonice? Parenchyma redux. Antiquity 87:971–984.

Puech PF. 1999. Usures des dents et sédentarisation? Doss Pour la Sci 22:90.

de Puydt M, Lohest M. 1887. L’homme contemporain du Mammouth à Spy (Namur). In: de

Radiguès de Chennevière H, editor. Fédération archéologique et historique de
Belgique. Namur. p 205–240.

R Core Team. 2014. R: A language and environment for statistical computing. R Found Stat
Comput Vienna, Austria:URL http://www.R-project.org/.

Reinhard KJ, Mendonça de Souza SMF, Rodrigues C, Kimmerle E, Dorsey-Vinton S. 2001.

Microfossils in dental calculus: a new perspective on diet and dental disease. In:
Williams E, editor. Human Remains: Conservation, Retrieval and Analysis. London:
British Archaeology Research Council. p 113–118.

Revedin A, Aranguren B, Becattini R, Longo L, Marconi E, Lippi MM, Skakun N, Sinitsyn A,

Spiridonova E, Svoboda J. 2010. Thirty thousand-year-old evidence of plant food
processing. Proc Natl Acad Sci U S A 107:18815–9.

Richards MP, Pettitt PB, Stiner MC, Trinkaus E. 2001. Stable isotope evidence for increasing
dietary breadth in the European mid-Upper Paleolithic. Proc Natl Acad Sci U S A
98:6528–6532.

Richards MP, Pettitt PB, Trinkaus E, Smith FH, Paunović M, Karavanić I. 2000. Neanderthal
diet at Vindija and Neanderthal predation: the evidence from stable isotopes. Proc Natl
Acad Sci U S A 97:7663–7666.

Richards MP, Schmitz RW. 2008. Isotope evidence for the diet of the Neanderthal type
specimen. Antiquity 82:553–559.

155
Richards MP, Taylor G, Steele TE, McPherron SP, Soressi M, Jaubert J, Orschiedt J, Mallye JB,
Rendu W, Hublin J-J. 2008. Isotopic dietary analysis of a Neanderthal and associated
fauna from the site of Jonzac (Charente-Maritime), France. J Hum Evol 55:179–185.

Richards MP, Trinkaus E. 2009. Out of Africa: modern human origins special feature:
isotopic evidence for the diets of European Neanderthals and early modern humans.
Proc Natl Acad Sci U S A 106:16034–9.

Rink WJ, Schwarcz HP, Valoch K, Seitl L, Stringer CB. 1996. ESR Dating of Micoquian
Industry and Neanderthal Remains at Kůlna Cave, Czech Republic. J Archaeol Sci
23:889–901.

Roberts-Harry EA, Clerehugh V. 2000. Subgingival calculus: where are we now? A

comparative review. J Dent 28:93–102.

Roebroeks W, Hublin J-J, Macdonald K. 2011. Continuities and Discontinuities in Neandertal

Presence: A Closer Look at Northwestern Europe.

Roger T, Darlas A. 2008. Upper-Pleistocene bird remains from Kalamakia Cave. In: Darlas A,
Milhailović D, editors. The Palaeolithic of the Balkans. Proceedings of the XV UISPP
World Congress. Oxford: BAR International Series 1819. p 69–76.

Rosen A. 2010. Natufian plant exploitation: Managing risk and stability in an environment of
change. Eurasian Prehistory 7:117–131.

Rosner A. 2011. Roasting Green Wheat in Galilee. Gastronomica 11:60–68.

Ross C. 1998. Primate Life Histories. Evol Anthropol 6:54–63.

Rothman JM, Van Soest PJ, Pell AN. 2006. Decaying wood is a sodium source for mountain
gorillas. Biol Lett 2:321–4.

Rougier H, Crevecoeur I, Beauval C, Bocherens H, Flas D, Germonpré M, Semal P, Van der

Plicht J. 2012. New data from an old site : Neandertals at Goyet (Belgium) and their
mortuary behavior. Am J Phys Anthropol 147:252–253.

Rougier H, Crevecoeur I, Beauval C, Flas D, Bocherens H, Wißing C, Germonpré M, Semal

P, van der Plicht J. 2014. New fossils at the “Troisième Caverne” of Goyet (Belgium)
and the mortuary practises of late Neandertals. In: Middle Palaeolithic in North West
Europe: Multidisciplinary Approaches. Namur. p 35.

Ruebens K, McPherron SJP, Hublin J-J. 2015. On the local Mousterian origin of the
Châtelperronian: Integrating typo-technological, chronostratigraphic and contextual
data. J Hum Evol.

Salazar-García DC. 2012. Isótopos, dieta y movilidad en el País Valenciano. Aplicación a

restos humanos del Paleolítico medio al Neolítico final.

Salazar-García DC, Power RC, Sanchis Serra A, Villaverde V, Walker MJ, Henry AG. 2013.
Neanderthal diets in central and southeastern Mediterranean Iberia. Quat Int 318:3–18.

156
Salazar-García DC, Richards MP, Nehlich O, Henry AG. 2014. Dental calculus is not
equivalent to bone collagen for isotope analysis: a comparison between carbon and
nitrogen stable isotope analysis of bulk dental calculus, bone and dentine collagen from
same individuals from the Medieval site of El Raval (Alicante. J Archaeol Sci 47:70–77.

Samuel D. 1996. Investigation of Ancient Egyptian Baking and Brewing Methods by

Correlative Microscopy. Science (80- ) 273:488–490.

Sanchis A. 2012. Los lagomorfos del Paleolítico medio en la vertiente mediterránea ibérica.
Diputación De Valencia.

Sandgathe DM, Hayden B. 2003. Did Neanderthals eat inner bark? Antiquity 77:709–718.

Sankararaman S, Patterson N, Li H, Pääbo S, Reich D. 2012. The date of interbreeding

between Neandertals and modern humans. PLoS Genet 8:e1002947.

Santos JL, Saus E, Smalley S V., Cataldo LR, Alberti G, Parada J, Gratacòs M, Estivill X. 2012.
Copy number polymorphism of the salivary amylase gene: implications in human
nutrition research. J Nutrigenet Nutrigenomics 5:117–31.

Saul H, Wilson J, Heron CP, Glykou A, Hartz S, Craig OE. 2012. A systematic approach to
the recovery and identification of starches from carbonised deposits on ceramic vessels.
J Archaeol Sci 39:3483–3492.

Savard M, Nesbitt M, Jones MK. 2006. The role of wild grasses in subsistence and sedentism:
new evidence from the northern Fertile Crescent. World Archaeol 38:179–196.

Schoch WH, Bigga G, Böhner U, Richter P, Terberger T. 2015. New insights on the wooden
weapons from the Paleolithic site of Schöningen. J Hum Evol 89:214–225.

Schwarcz HP, Schoeninger MJ. 1991. Stable isotope analyses in human nutritional ecology.
Am J Phys Anthropol 34:283–321.

Scott RS, Teaford MF, Ungar PS. 2012. Dental microwear texture and anthropoid diets. Am J
Phys Anthropol 147:551–79.

Selwitz RH, Ismail AI, Pitts NB. 2007. Dental caries. Lancet 369:51–9.

Semal P, Rougier H, Crevecoeur I, Jungels C, Flas D, Hauzeur A, Maureille B, Germonpré M,

Bocherens H, Pirson S, Cammaert L, De Clerck N, Hambucken A, Higham T, Toussaint
M, Van der Plicht J. 2009. New data on the late Neandertals: Direct dating of the
Belgian Spy fossils. Am J Phys Anthropol 138:421–428.

Sergi S. 1954. La mandibola neandertaliana Circeo II. Riv Di Antropol 41:305–344.

Simms SR. 1985. Acquisition Cost and Nutritional Data on Great Basin Resources. J Calif Gt
Basin Anthropol 7:117–126.

Simopoulos AP. 2004. Omega-3 Fatty Acids and Antioxidants in Edible Wild Plants. Biol Res
37:263–277.

Sistiaga A, Mallol C, Galván B, Summons RE. 2014. The Neanderthal meal: a new

157
 perspective using faecal biomarkers. PLoS One 9:e101045.

Slimak L, Svendsen JI, Mangerud J, Plisson H, Heggen HP, Brugere A, Pavlov PY. 2011. Late
Mousterian Persistence near the Arctic Circle. Science (80- ) 332:841–845.

Smith CI, Chamberlain AT, Riley MS, Cooper A, Stringer CB, Collins MJ. 2001. Neanderthal
DNA: Not just old but old and cold? Nature 410:771–772.

Smith EA. 2011. Endless forms: human behavioural diversity and evolved universals. Philos
Trans R Soc London B Biol Sci 366:325–332.

Smith FH, Boyd DC, Malez M. 1985. Additional upper pleistocene human remains from
Vindija cave, Croatia, Yugoslavia. Am J Phys Anthropol 68:375–383.

Smith TM, Toussaint M, Reid DJ, Olejniczak AJ, Hublin J-J. 2007. Rapid dental development
in a Middle Paleolithic Belgian Neanderthal. Proc Natl Acad Sci U S A 104:20220–5.

Snodgrass JJ, Leonard WR. 2009. Neandertal energetics revisited: insights into population
dynamics and life history evolution. PaleoAnthropology:220−237.

Solecki RS. 1960. Three adult Neanderthal skeletons from Shanidar cave, northern Iraq.
Annu Rep Board Regents Smithson Inst:603–635.

Solecki RS. 1961. New anthropological discoveries at Shanidar, northern Iraq. Trans N Y
Acad Sci 23:690–699.

Sołtysiak A. 2012. Comment: low dental caries rate in Neandertals: the result of diet or the
oral flora composition? Homo Int Zeitschrift für die vergleichende Forsch am
Menschen 63:110–3.

Sorensen M V., Leonard WR. 2001. Neandertal energetics and foraging efficiency. J Hum
Evol 40:483–95.

Speth JD. 2010. The Paleoanthropology and Archaeology of Big-Game Hunting. New York,
NY: Springer New York.

Speth JD, Clark JL. 2006. Hunting and overhunting in the Levantine Late Middle
Palaeolithic. Before Farming 3:1–42.

Speth JD, Spielmann KA. 1983. Energy source, protein metabolism, and hunter-gatherer
subsistence strategies. J Anthropol Archaeol 2:1–31.

Speth JD, Tchernov E. 2001. Neandertal hunting and meat processing in the Near East,
Evidence from Kebara cave (Israel). In: Stanford C, Bunn H, editors. Meat-eating and
Human evolution. Oxford: Oxford University Press. p 52–72.

Spoor F, Hublin J-J, Braun M, Zonneveld F. 2003. The bony labyrinth of Neanderthals. J
Hum Evol 44:141–165.

Squires BT. 1953. Human salivary amylase secretion in relation to diet. J Physiol 119:153–156.

Sroubek P, Diehl JF, Kadlec J, Valoch K. 2001. A Late Pleistocene palaeoclimate record based

158
 on mineral magnetic properties of the entrance facies sediments of Kulna Cave, Czech
Republic. Geophys J Int 147:247–262.

Ståhlberg S, Svanberg I. 2010. Gathering Food from Rodent Nests in Siberia. J Ethnobiol
30:184–202.

Ståhlberg S, Svanberg I. 2012. Gathering dog’s tooth violet (Erythronium sibiricum) in Siberia.
J la Soc Finno-Ougrienne 91:349–351.

Stahler DR, Smith DW, Guernsey DS. 2006. Foraging and feeding ecology of the gray wolf
(Canis lupus): lessons from Yellowstone National Park, Wyoming, USA. J Nutr
136:1923S–1926S.

Stefansson V. 1956. The Fat of the Land. New York: The Macmillan Company.

Stewart JR. 2005. The ecology and adaptation of Neanderthals during the non-analogue
environment of Oxygen Isotope Stage 3. Quat Int 137:35–46.

Stiner MC. 1994. Honor among thieves: a zooarchaeological study of Neandertal ecology.
Princeton: Princeton University Press.

Stiner MC. 1999. Paleolithic Population Growth Pulses Evidenced by Small Animal
Exploitation. Science (80- ) 283:190–194.

Stiner MC. 2001. Thirty years on the “Broad Spectrum Revolution” and paleolithic
demography. Proc Natl Acad Sci U S A 98:6993–6996.

Stiner MC. 2013. An Unshakable Middle Paleolithic? Trends versus Conservatism in the
Predatory Niche and Their Social Ramifications. Curr Anthropol 54:S288–S304.

Stiner MC, Kuhn SL. 1992. Subsistence, Technology, and Adaptive Variation in Middle
Paleolithic Italy. Am Anthropol 94:306–339.

Stiner MC, Kuhn SL. 2006. Changes in the “Connectedness” and Resilience of Paleolithic
Societies in Mediterranean Ecosystems. Hum Ecol 34:693–712.

Stiner MC, Kuhn SL. 2009. Paleolithic Diet and the Division of Labor in Mediterranean
Eurasia. In: Hublin J-J, Richards MP, editors. The Evolution of Hominin Diets:
Integrating Approaches to the Study of Palaeolithic Subsistence. Dordrecht: Springer. p
155–168.

Stiner MC, Munro ND. 2002. Approaches to Prehistoric Diet Breadth, Demography, and
Prey Ranking Systems in Time and Space. J Archaeol Method Theory 9:181–214.

Stiner MC, Munro ND, Surovell TA. 2000. The Tortoise and the Hare: Small Game Use, the
Broad Spectrum Revolution, and Paleolithic Demography. Curr Anthropol 41:39–79.

Stoneking M, Krause J. 2011. Learning about human population history from ancient and
modern genomes. Nat Rev Genet 12:603–14.

Straus LG. 1992. Iberia Before the Iberians: The Stone Age Prehistory of Cantabrian Spain.
Albuquerque: University of New Mexico Press.

159
Stringer CB, Barton RNE, Finlayson JC. 2000. Neanderthals on the edge: papers presented at
the conference marking the 150th anniversary of the Forbes’ Quarry discovery,
Gibralta. (Stringer CB, Barton RNE, Finlayson JC, editors.). Oxford: Oxbow.

Stringer CB, Finlayson JC, Barton RNE, Fernández-Jalvo Y, Cáceres I, Sabin RC, Rhodes EJ,
Currant AP, Rodríguez-Vidal J, Giles Pacheco F, Riquelme-Cantal JA. 2008.
Neanderthal exploitation of marine mammals in Gibraltar. Proc Natl Acad Sci U S A
105:14319–14324.

Szczęsna T. 2007. Concentration of selected elements in honeybee-collected pollen. J Apic Sci

51:5–13.

Szpak P. 2011. Fish bone chemistry and ultrastructure: implications for taphonomy and
stable isotope analysis. J Archaeol Sci 38:3358–3372.

Tao D, Zhang J, Zheng W, Cao Y, Sun K, Jin S. 2015. Starch grain analysis of human dental
calculus to investigate Neolithic consumption of plants in the middle Yellow River
Valley, China: A case study on Gouwan site. J Archaeol Sci Reports 2:485–491.

Tattersall I, Schwartz JH. 1999. Hominids and hybrids: The place of Neanderthals in human
evolution. Proc Natl Acad Sci U S A 96:7117–7119.

Teaford MF, Ungar PS. 2000. Diet and the evolution of the earliest human ancestors. Proc
Natl Acad Sci U S A 97:13506–13511.

Thieme H. 2000. Lower Palaeolithic Hunting Weapons From Schöningen, Germany - The
Oldest Spears in the World. Acta Anthropol Sin 19:136–143.

Thissen L, Özbal H, Türkekul Bıyık A, Gerritsen F, Özbal R. 2010. The land of milk?
Approaching dietary preferences of Late Neolithic communities in NW Anatolia.
Leiden J Pottery Stud 26:157–172.

Tillier A, Arensburg B, Rak Y, Vandermeersch B. 1995. Middle Palaeolithic dental caries:

new evidence from Kebara (Mount Carmel, Israel). J Hum Evol 29:189–192.

Toepfer V. 1958. Steingeräte und Palökologie der mittelpaläolithischen Fundstelle Rabutz

bei Halle (Saale). Jahresschrift für Mitteldeutsche Vor 41/42:140–177.

Tomczyk J. 2012. Comments on Soltysiak’s paper: “Comment: Low dental caries rate in
Neandertals: The result of diet or the oral flora compositions?” HOMO - J Comp Hum
Biol 63:311–314.

Torrence R, Barton H. 2006. Ancient Starch Research. (Torrence R, Barton H, editors.).

Walnut Creek: Left Coast Press.

Torrence R, Wright R, Conway R. 2004. Identification of starch granules using image

analysis and multivariate techniques. J Archaeol Sci 31:519–532.

Toussaint M. 2006. Research in the caves of Goyet (Gesves, Province of Namur, Belgium). In:
Demarsin B, Otte M, editors. Neanderthals in Europe. Liege. p 115–134.

160
Trinkaus E. 1981. Neanderthal limb proportions and cold adaptation. In: Stringer CB, editor.
Aspects of Human Evolution. London: Taylor and Francis. p 187–224.

Trinkaus E, Walker MJ. The People of Palomas: The Neandertals from the Sima de las
Palomas del Cabezo Gordo, Southeastern Spain. College Station: Texas A & M
University Press.

Tromp M. 2012. Large-scale Analysis of Microfossils Extracted from Human Rapanui Dental
Calculus: a Dual-Method Approach Using SEMEDS and Light Microscopy to Address
Ancient Dietary Hypotheses.

Tromp M, Dudgeon J V. 2015. Differentiating dietary and non-dietary microfossils extracted

from human dental calculus: the importance of sweet potato to ancient diet on Rapa
Nui. J Archaeol Sci 54:54–63.

Tsartsidou G, Lev-Yadun S, Albert RM, Miller-Rosen A, Efstratiou N, Weiner S. 2007. The

phytolith archaeological record: strengths and weaknesses evaluated based on a
quantitative modern reference collection from Greece. J Archaeol Sci 34:1262–1275.

Ungar PS. 1998. Dental allometry, morphology, and wear as evidence for diet in fossil
primates. Evol Anthropol Issues, News, Rev 6:205–217.

Ungar PS, Grine FE, Teaford MF. 2008. Dental microwear and diet of the Plio-Pleistocene
hominin Paranthropus boisei. PLoS One 3:e2044.

Ungar PS, Sponheimer M. 2011. The Diets of Early Hominins. Science (80- ) 334:190–193.

Valensi P, Psathi E. 2004. Faunal Exploitation duringtheMiddle Palaeolithic in South-eastern

France and North-western Italy. Int J Osteoarchaeol 14:256–272.

Valoch K. 1970. Early Middle Palaeolithic (Stratum 14) in the Kulna Cave near Sloup in the
Moravian Karst (Czechoslovakia). World Archaeol 2:28–38.

Van der Veen M. 2007. Formation processes of desiccated and carbonized plant remains –
the identification of routine practice. J Archaeol Sci 34:968–990.

Venables WN, Ripley BD. 2002. Modern Applied Statistics with S. 4th Editio. New York:
Springer.

Vernot B, Akey JM. 2015. Complex History of Admixture between Modern Humans and
Neandertals. Am J Hum Genet 96:448–453.

Verpoorte A. 2006. Neanderthal energetics and spatial behaviour. Before Farming 3:1–6.

Verpoorte A. 2009. Limiting factors on early modern human dispersals: The human
biogeography of late Pleniglacial Europe. Quat Int 201:77–85.

Villa P, Soriano S. 2010. Hunting weapons of Neanderthals and early modern humans in
South Africa: similarities and differences. J Anthropol Res 66:5–38.

Wadley L, Langejans GH. 2014. Preliminary Study Of Scrapers Around Combustion

Features In Layer Ss, Sibudu, 58000 Years Ago. South African Archaeol Bull 69:19–33.

161
Wales N. 2012. Modeling Neanderthal clothing using ethnographic analogues. J Hum Evol
63:781–95.

Walker M, Ortega J. 2011. Morphology, body proportions, and postcranial hypertrophy of a

female Neandertal from the Sima de las Palomas, southeastern Spain. Proc Natl Acad
Sci 108:10087–10091.

Walker MJ, Gibert J, López M V., Lombardi AV, Pérez-Pérez A, Zapata J, Ortega J, Higham
T, Pike A, Schwenninger J-L, Zilhão J, Trinkaus E. 2008. Late neandertals in
southeastern Iberia: Sima de las Palomas del Cabezo Gordo, Murcia, Spain. Proc Natl
Acad Sci U S A 105:20631–6.

Walker MJ, Gibert J, Sánchez F, Lombardi AV, Serrano I, Gómez A, Eastham A, Ribot F,
Arribas A, Cuenca A, Gibert L, Albadalejo S, Andreu JA. 1999. Excavations at new sites
of early man in Murcia: Sima de las Palomas del Cabezo Gordo and Cueva Negra del
Estrecho del Río Quípar de la Encarnación. Hum Evol 14:99–123.

Walker MJ, Lombardi AV, Zapata J, Trinkaus E. 2010. Neandertal mandibles from the Sima
de las Palomas del Cabezo Gordo, Murcia, southeastern Spain. Am J Phys Anthropol
142:261–72.

Walker MJ, López-Martínez, M. V., Ortega-Rodrigáñez, J. Haber-Uriarte M, López-Jiménez

A, Avilés-Fernández A, Polo-Camacho, J. L., Campillo-Boj M, García-Torres J, Carrión-
García JS, San Nicolás del Toro, M., Rodríguez-Estrella T. 2012. The excavation of
buried articulated Neanderthal skeletons at Sima de las Palomas (Murcia, SE Spain),.
Quat Int 259:7–21.

Walker MJ, Ortega J, López M V., Parmová K, Trinkaus E. 2011a. Neandertal postcranial
remains from the Sima de las Palomas del Cabezo Gordo, Murcia, southeastern Spain.
Am J Phys Anthropol 144:505–15.

Walker MJ, Zapata J, Lombardi A V., Trinkaus E. 2011b. New evidence of dental pathology
in 40,000-year-old Neandertals. J Dent Res 90:428–32.

Vander Wall SB. 2001. The evolutionary ecology of nut dispersal. Bot Rev 67:74–117.

Warinner C, Rodrigues JFM, Vyas R, Trachsel C, Shved N, Grossmann J, Radini A, Hancock

Y, Tito RY, Fiddyment S, Speller C, Hendy J, Charlton S, Luder HU, Salazar-García DC,
Eppler E, Seiler R, Hansen LH, Castruita JAS, Barkow-Oesterreicher S, Teoh KY,
Kelstrup CD, Olsen J V., Nanni P, Kawai T, Willerslev E, von Mering C, Lewis CM,
Collins MJ, Gilbert MTP, Rühli F, Cappellini E. 2014. Pathogens and host immunity in
the ancient human oral cavity. Nat Genet 46:336–44.

Watts DP, Mitani JC. 2015. Hunting and Prey Switching by Chimpanzees (Pan troglodytes
schweinfurthii) at Ngogo. Int J Primatol 36:728–748.

Weiss E, Wetterstrom W, Nadel D, Bar-Yosef O. 2004. The broad spectrum revisited:

evidence from plant remains. Proc Natl Acad Sci U S A 101:9551–5.

Weyrich LS, Dobney K, Cooper A. 2015. Ancient DNA analysis of dental calculus. J Hum

162
 Evol 79:119–124.

Wiessner P. 2003. Owners of the future: Calories, cash, and self-sufficiency in the Nyae Nyae
area between 1996 and 2003. Vis Anthropol Rev 19:149–159.

Wild EM, Paunović M, Rabeder G, Steffan I, Steier P. 2001. Age determination of fossil bones
from the Vindija Neanderthal site in Croatia. Radiocarbon 43:1021–1028.

Wilson J, Hardy K, Allen R, Copeland L, Wrangham RW, Collins MJ. 2010. Automated
classification of starch granules using supervised pattern recognition of morphological
properties. J Archaeol Sci 37:594–604.

Winterhalder B. 1986. Diet choice, risk, and food sharing in a stochastic environment. J
Anthropol Archaeol 5:369–392.

Winterhalder B, Smith EA. 2000. Analyzing adaptive strategies: Human behavioral ecology
at twenty-five. Evol Anthropol 9:51–72.

Wißing C, Rougier H, Crevecoeur I, Germonpré M, Naito YI, Semal P, Bocherens H. 2015.

Isotopic evidence for dietary ecology of late Neandertals in North-Western Europe.
Quat Int.

Wolpoff MH, Smith FH, Malez M, Radovčić J, Rukavina D. 1981. Upper pleistocene human
remains from Vindija cave, Croatia, Yugoslavia. Am J Phys Anthropol 54:499–545.

Wood RE, Barroso-Ruíz C, Caparrós M, Jordá Pardo JF, Galván Santos B, Higham TFG.
2013a. Radiocarbon dating casts doubt on the late chronology of the Middle to Upper
Palaeolithic transition in southern Iberia. Proc Natl Acad Sci U S A 110:2781–6.

Wood RE, Higham TFG, De Torres T, Tisnérat-Laborde N, Valladas H, Ortiz JE, Lalueza-Fox
C, Sánchez-Moral S, Cañaveras JC, Rosas A, Santamaría D, de la Rasilla M. 2013b. A
New Date for the Neanderthals from El Sidrón Cave (Asturias, Northern Spain).
Archaeometry 55:148–158.

Wrangham RW. 2000. Why are male chimpanzees more gregarious than mothers? In:
Kappeler P, editor. Primate Males: Causes and Consequences of Variation in Group
Composition. Cambridge: Cambridge University Press. p 248–258.

Wrangham RW, Smuts BB. 1980. Sex differences in the behavioural ecology of chimpanzees
in the Gombe National Park, Tanzania. J Reprod Fertil Suppl Suppl 28:13–31.

Wright KI. 1994. Ground-Stone Tools and Hunter-Gatherer Subsistence in Southwest Asia:
Implications for the Transition to Farming. Am Antiq 59:238–263.

Xia J, Zheng J, Huang D, Tian ZR, Chen L, Zhou Z, Ungar PS, Qian L. 2015. New model to
explain tooth wear with implications for microwear formation and diet reconstruction.
Proc Natl Acad Sci U S A 112:10669–10672.

Yellen JE. 1991a. Small mammals: !Kung San utilization and the production of faunal
assemblages. J Anthropol Archaeol 10:1–26.

163
Yellen JE. 1991b. Small mammals: Post-discard patterning of !Kung San faunal remains. J
Anthropol Archaeol 10:152–192.

Yesner DR. 1989. Moose hunters of the boreal forest? A re-examination of subsistence
patterns in the western subarctic. Arctic 42:97–108.

Young TK. 1996. Obesity, Central Fat Patterning, and Their Metabolic Correlates among the
Inuit of the Central Canadian Arctic. Hum Biol 68:245–263.

El Zaatari S. 2007. Ecogeographic variation in Neandertal dietary habits: evidence from

microwear texture analysis.

El Zaatari S. 2010. Occlusal microwear texture analysis and the diets of historical/prehistoric
hunter-gatherers. Int J Osteoarchaeol 20:67–87.

El Zaatari S, Grine FE, Ungar PS, Hublin J-J. 2011. Ecogeographic variation in Neandertal
dietary habits: evidence from occlusal molar microwear texture analysis. J Hum Evol
61:411–24.

El Zaatari S, Grine FE, Ungar PS, Hublin J-J. 2016. Neandertal versus Modern Human
Dietary Responses to Climatic Fluctuations. PLoS One 11:e0153277.

El Zaatari S, Hublin J-J. 2014. Diet of upper paleolithic modern humans: evidence from
microwear texture analysis. Am J Phys Anthropol 153:570–81.

Zeder MA. 2012. The Broad Spectrum Revolution at 40: Resource diversity, intensification,
and an alternative to optimal foraging explanations. J Anthropol Archaeol 31:241–264.

Zhang J, Lu H, Huang L. 2014. Calciphytoliths (calcium oxalate crystals) analysis for the
identification of decayed tea plants (Camellia sinensis L.). Sci Rep 4:6703.

Zilhão J, Angelucci DE, Badal-García E, D’Errico F, Daniel F, Dayet L, Douka K, Higham

TFG, Martínez-Sánchez MJ, Montes-Bernárdez R, Murcia-Mascarós S, Pérez-Sirvent C,
Roldán-García C, Vanhaeren M, Villaverde V, Wood RE, Zapata J. 2010. Symbolic use
of marine shells and mineral pigments by Iberian Neandertals. Proc Natl Acad Sci U S
A 107:1023–8.

Zilhão J, d’Errico F, Bordes J-G, Lenoble A, Texier J-P, Rigaud J-P. 2006. Analysis of
Aurignacian interstratification at the Chatelperronian-type site and implications for the
behavioral modernity of Neandertals. Proc Natl Acad Sci U S A 103:12643–8.

Zimov SA, Zimov NS, Tikhonov AN, Chapin FS. 2012. Mammoth steppe: a high-
productivity phenomenon. Quat Sci Rev 57:26–45.

Zwyns N, Roebroeks W, McPherron SP, Jagich A, Hublin J-J. 2012. Comment on “Late
Mousterian Persistence near the Arctic Circle.” Science (80- ) 335:167–167.

164
 7

Appendixes

7.1 Chapter three appendix

Appendix tables

Appendix table 1: Elemental composition of standards from EDX.

No. Grouping C O Na Mg Al Si P Ca F N K S Cl Cr Mn

Fru.1 Fructose 90.8 9.17

Fru. 2 Fructose 92.2 7.78
Fru. 3 Fructose 90.4 9.58
Fru. 4 Fructose 91.4 8.62
Fru 5 Fructose 93.2 6.8
Suc. 1 Sucrose 90.3 9.75
Suc. 2 Sucrose 91.8 8.21
Suc. 3 Sucrose 89.9 10.1
Suc. 4 Sucrose 92.9 7.15
Suc. 5 Sucrose 92.4 7.58
Mal. 1 Maltose 60.7 39.3
Mal. 2 Maltose 62 38
Mal. 3 Maltose 58 42
Mal. 4 Maltose 62.8 37.2
Mal. 5 Maltose 58 42
Glu. 1 Glucose 62.3 37.7
Glu. 2 Glucose 57.9 42.1
Glu. 3 Glucose 57.9 42.1
Glu. 4 Glucose 59.1 40.9
Glu. 5 Glucose 60.1 39.9
Corn 1 Corn starch 58 42
Corn 2 Corn starch 61.7 38.3
Corn 3 Corn starch 61.6 38.4
Corn 4 Corn starch 59.2 40.8
Corn 5 Corn starch 62.1 37.9
Cola 3.2 Kola starch 81.8 14.3 14.9 0.3 0.5 0.2 0.17 0.3 0.67 1.4 0 0.25
Cafr-2 Kola starch 54.8 5.28 1.51 2.02 23 4 5 4.7
Cola 2.3 Kola starch 67.5 26.2 0.24 0.88 0.4 0.2 0.67 0.51 2.4 0.59 0
Cola 2.4 Kola starch 70 22.8 0.16 0.61 0.5 0.5 0.86 1.4 1.9 0.75 1
Cola 2.5 Kola starch 66.1 25.9 0.7 0.2 0.3 0.76 0.13 0.7 0
Xylia 1 Xylia starch 76.8 21.4 0.17 0.21 0.5 0.1 0.11 0.1 0.2 0.24 0
Xylia 2 Xylia starch 78.1 21.1 0.8
Xylia 3 Xylia starch 74.7 20.2 0.42 0.4 0.94 2.5 0.89

165
Xylia 4 Xylia starch 79.7 16.2 0.62 1 0.8 0.7 0.92
Xylia 5 Xylia starch 75.8 22.2 0.43 0.6 0.18 0.5 0.25
Pot. 1 Potato starch 82.8 17.2
Pot. 2 Potato starch 83.1 16.9
Pot. 3 Potato starch 84.3 15.7
Pot. 4 Potato starch 84.1 15.9
Pot. 5 Potato starch 82.1 16.5 1.4
wtfr-1 Wheat starch 86.2 13.8 6.9
wtfr-2 Wheat starch 86.1 13.9
wtfr-3 Wheat starch 89.9 10.1
wheat n Wheat starch 91 8.97
wheat n2 Wheat starch 90.2 9.81

Appendix table 2: Elemental composition of degraded and native starch.

No. Grouping C O Na Mg Al Si P F K S Cl Cr Mn
Cafr-1 Kola native 64.7 3.8 1.46 0.91 18 4.44 3.03 3.97
Cafr-2 Kola native 54.8 5.3 1.51 2.02 23 3.92 4.97 4.67
Cafr-3 Kola native 60.1 6.0 1.47 1.66 17 5.25 3.81 5.03
Cola 2 1 Kola native 65.3 29.6 5.1
Cola 2 2 Kola native 67.8 28.5 3.7
Cola 2 3 Kola native 67.5 26.2 0.24 0.88 0.35 0.24 0.67 0.51 2.4 0.59 0.46
Cola 3 1 Kola native 80.8 15.3 0.32 0.56 0.3 0.08 0.21 0.55 1.5
Cola 3 2 Kola native 81.8 14.3 0.3 0.51 0.15 0.17 0.27 0.67 1.4 0.35 0.15
Cola 3 3 Kola native 77.5 18.0 0.28 0.62 0.26 0.18 0.43 0.47 1.8 0.3 0.13
Csfr-1 1 Gabon nut native 75.1 24.9
Csfr-1 2 Gabon nut native 94.9 5.1
Csfr-1 3 Gabon nut native 94.9 5.1
Csfr-2 1 Gabon nut native 74.5 20.3
Csfr-2 2 Gabon nut native 69.2 25.4
Csfr-2 3 Gabon nut native 68.4 24.8
Csfr-3 1 Gabon nut native 65.4 28.9
Csfr-3 2 Gabon nut native 62.6 33.7
Csfr-3 3 Gabon nut native 65.1 30.9
Wtfr-1 1 Wheat native 86.2 13.8
Wtfr-1 2 Wheat native 86.1 13.9
Wtfr-1 3 Wheat native 89.9 10.1
Wtfr-2 1 Wheat native 67.4 32.6
Wtfr-2 2 Wheat native 76.7 23.3
Wtfr-2 3 Wheat native 73.1 26.2
Wtfr-3 1 Wheat native 67.9 32.1
Wtfr-3 2 Wheat native 68.5 31.5
Wtfr-3 3 Wheat native 68.9 31.1
Ca301-1 Kola 30 mins 84.4 15.6
Ca301-2 Kola 30 mins 85.6 14.4
Ca301-3 Kola 30 mins 86.6 13.4

166
Ca302-1 Kola 30 mins 86.9 13.1
Ca302-2 Kola 30 mins 88.4 11.6
Ca302-3 Kola 30 mins 89.9 10.1
Ca303-1 Kola 30 mins 88.5 11.6
Ca303-2 Kola 30 mins 87.8 12.2
Ca303-3 Kola 30 mins 92.9 7.1
Cs301-1 Gabon nut 30 mins 90.5 9.5
Cs301-2 Gabon nut 30 mins 90.7 9.3
Cs301-3 Gabon nut 30 mins 86.7 13.3
Cs302-1 Gabon nut 30 mins 88.6 11.4
Cs302-2 Gabon nut 30 mins 87.6 12.4
Cs302-3 Gabon nut 30 mins 89.6 10.4
Cs303-1 Gabon nut 30 mins 90.0 10.0
Cs303-2 Gabon nut 30 mins 87.6 12.4
Cs303-3 Gabon nut 30 mins 89.6 10.4
Wt301-1 Wheat 30 mins 88.9 11.1
Wt301-2 Wheat 30 mins 85.7 14.3
Wt301-3 Wheat 30 mins 87.3 12.7
Wt302-1 Wheat 30 mins 84.2 15.9
Wt302-2 Wheat 30 mins 84.2 15.8
Wt302-3 Wheat 30 mins 87.6 12.4
Wt303-1 Wheat 30 mins 88.3 11.7
Wt303-2 Wheat 30 mins 87.0 13.0
Wt303-3 Wheat 30 mins 88.3 11.8
Ca901-1 Kola 90 mins 86.7 13.3
Ca901-2 Kola 90 mins 86.8 13.2
Ca901-3 Kola 90 mins 89.7 10.3
Ca902-1 Kola 90 mins 86.6 13.5
Ca902-2 Kola 90 mins 91.6 8.4
Ca902-3 Kola 90 mins 90.0 10.0
Ca903-1 Kola 90 mins 84.3 15.7
Ca903-2 Kola 90 mins 90.0 10.0
Ca903-3 Kola 90 mins 89.3 10.7
Cs901-1 Gabon nut 90 mins 90.6 9.4
Cs901-2 Gabon nut 90 mins 89.0 11.0
Cs901-3 Gabon nut 90 mins 84.9 15.1
Cs902-1 Gabon nut 90 mins 88.0 12.0
Cs902-2 Gabon nut 90 mins 89.6 10.4
Cs902-3 Gabon nut 90 mins 92.5 7.5
Cs903-1 Gabon nut 90 mins 86.0 14.0
Cs903-2 Gabon nut 90 mins 88.4 11.6
Cs903-3 Gabon nut 90 mins 90.5 9.5
Wt901-1 Wheat 90 mins 86.5 13.5
Wt901-2 Wheat 90 mins 86.1 13.9
Wt901-3 Wheat 90 mins 85.8 14.2
Wt902-1 Wheat 90 mins 82.8 17.2

167
Wt902-2 Wheat 90 mins 85.4 14.6
Wt902-3 Wheat 90 mins 88.7 11.3
Wt903-1 Wheat 90 mins 84.0 16.0
Wt903-2 Wheat 90 mins 84.8 15.3
Wt903-3 Wheat 90 mins 84.93 15.07

Appendix table 3: Elemental composition of calculus and microremains in calculus from EDX. T=Taï Forest
Chimpanzee, C=Camino de Molino.
Type

No. Category C O Na Mg Al Si P Ca F N K Ba La T

T Venus exposed starch 95.1 5.0

clump
T Venus starch clump 61.5 5.0 16.3 7.1 10.2
mantel
T Castor calculus matrix 14.2 9.9 0.5 0.63 0.5 0.8 22.8 50.6
1a T Castor starch cluster 63.1 8.2 1.4 1.87 0.8 0.7 12.3 11.7
1b T Castor starch cluster 60.1 6.4 1.4 1.62 1.2 2.3 13.4 13.8
3 T Castor unknown 13.0 14.4 0.7 0.61 0.2 0.1 25.4 45.6
microf
4 T Castor unknown 16.1 10.4 0.7 0.78 0.4 0.3 24.1 47.3
microf
5 T Castor unknown 25.3 10.9 0.6 0.71 0.5 0.6 19.8 41.6
microf
1 T Fanny unknown 13.7 10.2 0.4 0.59 1.0 1.1 22.7 50.3
microf
2 T Fanny unknown 5.1 19.2 0.2 7.12 17.9 35.1 4.8 10.6
particle
3 T Fanny unknown 9.4 8.7 0.5 0.63 0.6 0.6 24.3 55.2
particle
4 T Fanny unknown 26.2 9.0 0.7 0.67 1.2 0.5 18.5 42.9
particle
5 T Fanny unknown 6.4 15.5 0.15 0.5 71.2 2.0 4.3
particle
6 T Fanny unknown 61.3 4.3 0.2 0.29 11 2.7 9.9 21.3
microf
7 T Fanny unknown 79.9 7.1 0.6 0.1 0.2 3.9 2.6 5.6
microf
8 T Fanny unknown 10.4 3.3 0.2 0.61 0.4 0.1 23.5 61.5
microf
9 T Fanny unknown 47.3 7.8 0.4 0.8 0.4 0.5 11 31.8
microf
12 T Fanny phytolith 16.3 14.2 1.2 1 0.4 31.1 11.8 23.9
16 T Fanny unknown 46.2 8.0 0.7 1.13 1.3 1.5 17.2 24.0
microf
17 T Fanny unknown 7.2 4.5 0.47 3.2 6.3 13.7 64.7
microf
T Goma calculus matrix 6.8 9.9 0.3 1.16 2.0 2.6 24.0 51.2 0.1 1.6 0.5
1 T Goma phytolith 6.3 15.9 0.6 3.88 7.8 15 17.0 27.0 0.7 1.0 5
2 T Goma microremain 44.4 5.3 1 1.16 0.3 1 19.7 22.6 3.2 1.4
3 T Goma diatom 3.4 9.7 0.1 0.28 2.3 79.3 0.7 2.4 1.1 0.7
5 T Goma microremain 3.5 5.8 0.4 1.66 6.8 8.2 11.0 59.0 0.2 2.2 1.4
11 T Goma microremain 7.7 15.1 0.7 2.45 0.4 36.5 34.8 0.8 1.6
T Leo calculus matrix 6.9 11.0 0.4 1.53 0.6 1.2 26.4 51.9
1 T Leo microremain 38 9.3 0.4 0.62 12.1 18.1 8.0 13.5

168
 3 T Leo palm phytolith 7.1 20.1 7.8
11 T Leo invertebrate 91.7 8.3
15 T Leo microremain 8.0 9.5 0.8 2.21 2.3 3.7 9.5 61.0 1 2
18 T Leo unknown 21.6 78.4
T Rubra unknown 13.4 17.2 0.4 1.74 26.4 41.0
1 T Rubra unknown 8.7 18.1 0.54 26.2 39.8 6.7
2 T Rubra unknown 5.1 13.8 0.7 2.35 28.6 48.7
4 T Rubra diatom 41.0 11.9 0.1 0.28 0.9 20.5 7.7 17.6
14 T Rubra starch cluster 5.0 14.7 0.4 1.15 0.9 35.6 15.8 26.6
15 T Rubra diatom 4.3 12.4 0.1 0.38 19.9 57.6 1.9 3.5
20 T Rubra diatom 4 21.5 0.1 0.11 28.5 43.1 0.7 2.0
C SJ-13-32_1 rectangle 5.6 12.1 0.8 0.76 1.8 4.6 17.1 57.9
C SJ-13-32_2 unknown 7.1 13.1 9.62 16.9 41.9 1.7 4.5 3.4 2
C SJ-13-32_7 unknown 7.8 5.2 0.4 2.8 3 3.1 77.8
C SJ-13-32_10 unknown 3.0 7.1 0.7 1 88.2
C SJ-13-32_12 spicule 2.3 5.0 0.46 1.4 4.3 85.8 0.9
C SJ-13-32_16 unknown 1.8 9.2 1.08 3.4 8.8 75.4 0.4
C SJ-13-32_18 unknown 6.9 17.2 0.61 5.5 18.8 2.7 44.3 4.1
C SJ-13-33 unknown 4.4 10.0 0.2 0.35 0.3 1 21.2 62.5
C SJ-13-36 phytolith 4.1 11.6 0.27 19.9 31.2 1.4 31.6
C SJ-13-39 -3 unknown 3.6 7.5 0.9 2.3 9.4 76.2
C SJ-13-39 -7 spicule 2.5 9.0 0.46 0.6 1.5 6.5 79.5
C SJ-13-39 -11 unknown 5.4 14.5 1.3 4.8 13.2 14.7 46.1

169
 7.2 Chapter four appendix

7.2.1 Study population

The chimpanzee calculus samples derive from the Taï Chimpanzee osteology
collection of 77 chimpanzees curated at the Max Planck Institute for Evolutionary
Anthropology (MPI-EVA) in Leipzig, Germany. The remains were collected with as
many details as possible on sex, age and cause of death. All Taï Forest material and
data collected complied with the requirements and guidelines of the Ministère de
l’Enseignement Supérieure et de la Recherche Scientifique of Côte d’Ivoire, and
adhered to its legal requirements. When possible we sampled chimpanzees who had
known life histories, and ideally with comprehensive dietary records. Much of the
observational data relate to chimpanzees that are not part of this osteology
collection. Dietary records vary from thousands of observations over a decade to a
limited number over the course of a single day. After death, these individuals were
interred for defleshing and then later exhumed. Some of the skeletal material was
cleaned using strong disinfectants before storage to minimise the risk of disease
transmission.

It has been noted that chimpanzees produce less salivary α-amylase than
humans, especially humans from agricultural societies that consume high levels of
starch (Perry et al., 2007). Thus, starch entering the chimpanzee mouth may be less
readily hydrolysed than in human groups, which may make it more likely for
starches to enter and preserve in chimpanzee dental calculus than in human dental
calculus. However, if this patterns occurs in our samples it is unclear and it cannot
testable with our data.

7.2.2 Collection of calculus samples

Occasionally, chimpanzee calculus showed substantial flecks of dark material

that did not resemble calculus and appeared to be sediment contamination.
Chimpanzee samples where sediment contamination was suspected were omitted.
All chimpanzee remains sampled are curated at the Max Planck Institute for
Evolutionary Anthropology in Leipzig, Germany. Samples from two chimpanzees
(Vanessa and 13438) were omitted from analysis because their age at death was not
recorded. A sample from a further chimpanzee (Loukoum) was omitted due to
surface adherents on the calculus. The calculus we chose for the final complete

170
analysis came from molars of 24 individuals (12 males and 12 females) ranging in
age from between 12 and 552 months (1 and 46 years) old (Table 8).

7.2.3 Taï Forest plant reference collection

A microremain reference collection with 119 plant species was built using the
most frequently consumed chimpanzee plant foods in the Taï forest (Appendix table
4). Taï Chimpanzees consume a particularly diverse range of foods. We collected
plant parts that were documented as a specific component of the diet (fruits, seeds,
piths, leaves, stems, bark, flowers, and roots.) We also include fungal fruiting bodies
known to be consumed. Effort was made to include other rainforest edible plants not
recorded as chimpanzee foods. Although our reference collection is not exhaustive, it
incorporates the most important plants foods of the Taï Chimps, achieving coverage
of 89 % of the total dietary observations. Plants collected in the Taï Forest were
immediately preserved onsite either by freezing or by drying in 15 or 50 ml
centrifuge tubes with silica gel (Roth- T858.1 and P077.1, Karlsruhe, Germany).
Additionally, we collected some plant material from the University of Leipzig
Botanical Garden (marked as fresh in Appendix table 4) and analysed this material
fresh for starch or dried for phytoliths. We did not make a reference collection for
unsilicified plant microremains, as these microremains are unlikely to be
nondiagnostic.

Starch was analysed by directly mounting finely sliced dry plant material on
slides with approximately 10 µl of distilled water and 10 µl of a 25 % glycerol
solution. Starches were observed at 200-640 x magnification using a Zeiss Axioscope.
Phytoliths were isolated from plant material by dissolving weighed dried plant
material in ≥65 % nitric acid with a heating block to expedite the reaction. Small
quantities of the oxidiser potassium chlorate were added to encourage the process.

In most chimpanzee foods we observed either very few starch grains or none
at all, suggesting quantities too negligible to be detected or a complete lack of starch
in the plant. Plants that produced negligible numbers of starches were not analysed
for the identification model, because they did not have enough starch grains to build
a reference set of 50 starches. We found phytoliths were common in many species,
but many morphotypes are poorly studied in morphometric studies and cannot be
easily described using the variables we chose for our model (e.g. hair cells,
epidermal, cylindroids, plates and tracheid phytoliths). These morphotypes were

171
found in a number of genera in the reference collection plant but only in low
numbers.

Plants that had few phytoliths were not included. Furthermore, if

microremains were found in parts of a plant that chimpanzees do not eat, the plants
were not included (e.g. starch from Beilschmedia mannii seed). Thirteen starch- and
seven phytolith-producing plants were selected for developing identification criteria.
We chose to measure or quantify several variables on 50 microremains per species,
focusing on variables that past studies have shown to be effective in distinguishing
among starches and phytoliths (Torrence et al., 2004; Fenwick et al., 2011). Our
variables include max length, max width, area, shape, surface regularity, the number
of echinate spines, length of longest cross axis, type, number and length of cracks,
number of facets and lamellae (Appendix table 6). If abundant starches or phytoliths
were recovered, their abundance was analysed in order to assess the expected starch
and phytolith contribution to dental calculus. Starch content was established by
combining previous nutritional content studies (Oyebade, 1973; N’guessan, 2012).
For species where this data was not available we assessed starch content per gram
dried plant material colourimetrically using an Amyloglucosidase/α-amylase
method with a Megazyme Total Assay Kit (AA/AMG 11/01, AOAC Method 996.11,
AACC Method 76.13, ICC Standard Method No. 168). Phytolith content was
estimated by calculating the total weight of sample left after nitric acid digestion.

7.2.4 Identification of microremains by classification

Statistical approaches are increasingly used for the study and classification of
microremains (Wilson et al., 2010; Fenwick et al., 2011; Saul et al., 2012; Zhang et al.,
2014; Coster and Field, 2015). A variety of approaches have been implemented in
past studies such as image analysis (Colliot et al., 1997), linear discrimination
(Torrence et al., 2004), and factor regression analysis by principal components
(Fenwick et al., 2011). We used random forest-based classification because it is
robust, non-parametric and easily accommodates both large number of variables and
categorical data. Using this approach, we can easily see the most important variables
that drive the differences among the microremain types. The most important
variables in our phytolith model include length and the number of spines (Appendix
table 14). In the starch random forest model, area and length were the most
important variables (Appendix table 14).

172
 7.2.5 Model design and formulae

We predicted that number of microremains should increase with age, and

might vary by sex. We tested this using a negative binomial regression, with
microremain count as the response, and age and sex as predictors, weighting each
observation by the weight of the calculus sample (see detailed methods below). We
ran separate tests for phytoliths, unsilicified remains and starches.

The models described in R terminology are as follows:

Microremain type count~ chimpanzee age + chimpanzee sex, weights=calculus

sample weight

Expressed as a mathematical formula, this analysis is written as follows:

𝑦𝑖 = 𝑁𝑒𝑔𝑏𝑖𝑛(𝜇𝑖 , 𝑘)

𝑙𝑜𝑔(𝜇𝑗 ) = 𝛽0 + 𝑋𝑗 𝛽𝑗 + 𝜀

where 𝛽0 = 0
𝑝
𝑙𝑜𝑔(𝜇𝑗 ) = 𝛽0 + ∑ [ 𝛽11𝑗 chimp_age𝑗 + 𝛽12𝑗 chimpanzee sex𝑗 ] + 𝜀𝑗
𝑗=1

where 𝛽0 = 0

We predicted that more frequently consumed plants should be highly

represented in the chimpanzee calculus. To test this, we used an observational
random effect Poisson model. The count of microremains (starches or phytoliths)
belonging to a particular genus was our response variable, and the fixed predictors
were: (a) minutes spent consuming each genus, and (b) chimpanzee age in months.
Sex was included as a control predictor, and both calculus sample weight and
successful identification rate of each genus were included as weights. We accounted
for the variation in production of microremains in different genera by using
microremains content as an offset. We used counts of each genus predicted to be
present with the total minutes spent consuming each genus. The chimpanzee

173
individual was included as a random slope term, while year of death, tooth and food
type were treated as random intercept terms

The models described in R terminology are as follows:

The observational feeding records model. Key: obs_id=observation id,

plant_id=Plant genus, death_year=year that chimpanzee died,
mr_content=Prevalence of starch in each plant species, wt=Milligrams in each
sample, class_rate=Rate of successful identification in this species.

Count of each plant species~mins+age+sex+(1|obs_id)+(1|plant_id)+(1|tooth)+

(1|chimp_name)+(1|death_year)+(0+mins|chimp_name)+(0+mins|tooth)+(0+mins|d
eath_year)+(0+age|plant_id)+(0+age|tooth)+offset(log(mr_content)),
weight=class_rate+ calculus samples weight

In mathematical notation, the models are written as follows:

𝑝
𝑙𝑜𝑔𝑒 (𝜆) = −𝑛𝜆 + 𝑙𝑜𝑔𝑒 (𝜆) ∑ [ 𝛽11𝑗 mins𝑗 + 𝛽12𝑗 age𝑗 + 𝛽13𝑗 sex𝑗 ) + 𝛽21𝑗 +𝑢11𝑗 )tooth𝑗
𝑗=1
+ (𝛽22𝑗 +𝑢12𝑗 )death_year𝑗 + (𝛽23𝑗 +𝑢13𝑗 )plant_id𝑗 + (𝛽24𝑗 +𝑢14𝑗 )age𝑗
𝑝
−∑ ln[𝛽11𝑗 mins𝑗 + 𝛽12𝑗 age𝑗 + 𝛽13𝑗 sex𝑗 ) + 𝛽21𝑗 +𝑢11𝑗 )tooth𝑗
𝑗=1
+ (𝛽22𝑗 +𝑢12𝑗 )death_year𝑗 + (𝛽23𝑗 +𝑢13𝑗 )plant_id𝑗 + (𝛽24𝑗 +𝑢14𝑗 )age𝑗 ] !
+ 𝑢01 + 𝑢02 + 𝑢03 + 𝑢04 + 𝑢05 + 𝜀𝑗

174
 45

40

35

30
Starches/mg

25

20

15

10

0
1982 1985 1988 1991 1994 1997 2000 2003 2006 2009

Appendix fig. 1: Starches per mg in each chimpanzee calculus sample and year of death. Starches/mg incudes
the possible starch microremain category. Treatment of the skeletal remains and year of chimpanzee death
does not predict variation of starches per mg.

175
 Sacoglottis
Coula

Treculia
Calpocalyx
Xylia
Cola

Eremospatha

Panda

Gilbertiodendron

Elaeis

Laccosperma

Sarcophrynium
Aframomum

Piper

Napoleona

0 50,000 100,000 150,000 200,000 250,000 300,000 350,000 400,000

Minutes spent consuming each genus

Appendix fig. 2: Chimpanzee plant foods, ranked by minutes consumed. Plants in random forest model are in
red and those that are not are in blue. Chart omits foods eaten for <40 minutes. Our sample includes plants
that are frequently consumed (e.g. Sacoglottis and Coula) as well as those less often eaten (e.g. Piper and
Napoleona).

176
Tables

Appendix table 4: Inventory of plants and fungi analysed in reference collection. x=no microremain found.
o=microremains found and used for identification model. 1=found but not used in classification model due to
their complex morphology, 2=found but not included as they are very rare, 3=found but only in parts that are
not eaten. Prep=preparation. d=dried, fn=frozen and fh=fresh.

Fruit pulp

Fruit pulp
Flower

Flower
Stem

Stem
Shell

Shell
Seed

USO

Seed

USO
Bark

Bark

Prep
Leaf

Leaf
Pith

Pith
Plant genus Plant species Starch Phytoliths

Aframomum exscapum x x d
(Sims) Hepper
Aframomum cereum X x d
(Hook.f.)
K.Schum.
Afzelia bella Harms 1 d

Agaricus bispourus x d
(J.E.Lange)
Emil J. Imbach
Anchomanes difformis (Bl.) x fn
Engl.
Antiaris toxicaria subsp. x 2 d
welwitschii
(Engl.)
C.C.Berg
Auricularia auricula-judae. x x d
(Bull.)
J.Schröt.
Beilschmiedia mannii 2 d
(Meisn.)
Benth. &
Hook.f.
Bombax buonopozense x d
P.Beauv.
Bombax ceiba L. x 2 fh

Calpocalyx Sp. o d

Calpocalyx aubrevillei x X d
Pellegr.
Canarium schweinfurtii x x fn
Engl.
Castanola paradoxa (Gilg) x x d
Schellenb.
Chrysophyllum taiense x x x X x x d
Aubrév. &
Pellegr.
Cola nitida (Vent) x x x 1 x x d,
Schott & Endl. fh
Cola heterophylla (P x x x 1 x x d
Beauv.)
Schott. &
Endl.
Cola laterita K x x d
Schum.
Cordia platythyrsa x x x x d
Baker
Coula edulis Baill. x x x 1 x 1 d

Dacryodes klainaea x x fn
(Pierre)
H.J.Lam
Desplatsia chrysochlamys x X d
(Mildbr. &
Burret)
Mildbr. &
Burret
Detarium senegalense x x d
J.F.Gmel.

177
Dialium aubrevillei x x X x d
Pellegr.
Dialium dinklagei x x d
Harms
Dichapetalum heudelotii x X d
(Planch.) Baill.
Dioscorea burkilliana x d
J.Miège
Diospyros chevalieri De x d
Wild.
Diospyros manii Hiern x X 1 d
Diospyros sanza minika A x d
Chev.
Diospyros soubreana X d
F.White
Drypetes aubrevillei x x x d
Léandri
Duboscia viridifolia x d
(K.Schum.)
Mildbr.
Duguetia staudtii (Engl. 3 3 d
& Diels)
Chatrou
Elaeis guineenis Jacq. x x O o o d,
fh
Entandrophragma angolense x x d
(Welw.) C.
DC.
Eremospatha macrocarpa o o d
H.Wendl.
Erythrophleum ivorensis x fn
A.Chev
Ficus barteri 1 d
Sprague
Ficus elastica Roxb. x 1 fh

Ficus elasticoides De x d
Wild
Ficus lutea Vahl x d

Ficus polita Vahl 1 d

Gilbertiodendron splendidum o o x x d
(Hutch. &
Diels) J.
Léonard
Glyphaea brevis x 3 d
(Spreng.)
Monach.
Grewia biloba x x x x d
(Bunge.)Hand.
Mazz.
Grewia malacocarpa x x d
Mast.
Guibourtia tessmannii x d
(Harms)
J.Léonard
Halopegia azurea x x X x x x d
(K.Schum.)
K.Schum.
Harungana madagascariensis Lam. x x fn
ex Poir.
Heisteria parvifolia Sm. x d
Hexalobus crispiflorus x fn
A.Rich
Hypselodelphys violacea (Ridl.) x 1 d
Milne-Redh
Irvingia gabonensis x x d
(Aubry-
Lecomte ex
O'Rorke) Baill.
Irvingia grandifolia x d
(Engl.) Engl.
Keayodendron bridelioides x d

178
 (Gilg &
Mildbr. ex
Hutch. &
Dalziel)
Leandri
Klainedoxa gabonensis 3 fn,
Pierre d
Laccosperma secundiflorum x x d
(P.Beauv.)
Kuntze
Laccosperma opacum Drude x x d

Landolphia dulcis (Sabine x x x fn

ex G.Don)
Pichon
Magnistipula butayei x d
DeWild
Mammea africana Sabine x x d

Manilkara obovata x x fn
(Sabine &
G.Don)
J.H.Hemsl.
Manniophyton fulvum x d
Müll.Arg.
Memecylon Sp. x fn

Musanga Sp. x 1 1 d

Myrianthus Sp. x fn

Myrianthus arboreus x fn
P.Beauv.
Napoleona leonensis o d

Napoleonaea vogelii Hook. x X x fh

& Planch
Nauclea diderrichii (De x d
Wild. &
T.Durand)
Merr.ill
Nauclea xanthoxylon x x d

Pachira cubensis x fh
(A.Robyns)
Fern.Alonso
Palisota barteri Hook.f. 2 2 x x x d

Palisota bracteosa x x d
C.B.Clarke
Palisota hirsuta x d
(Thunb.)
K.Schum.
Panda oleosa Pierre x o X x d

Parinari excelsea Sabine x x 1 x fn

Parkia bicolor x x fn
A.Chev.
Pentaclethra macrophylla x d
Benth
Pentaclethra macrophylla x x d
Benth
Pentadesma butyracea x fn,
Sabine d
Piper betle L. x x 1 1 fh

Piper guineense o o d
Schumach. &
Thonn.
Piper longum L. x x d

Piper arboreum x 1 fh
Aubl.
Piper ornatum x fh
N.E.Br.
Pouteria pierrei x x x x d
(A.Chev.)

179
 Baehni

Pseudospondias Sp. x x fn

Pseudospondias microcarpa x x d
Engl
Psychotria bacteriophila x x d
Valeton
Pycnanthus angolensis x d
(Welw.) Warb.
Raphia sudanica x d
A.Chev.
Rhodognaphalon brevicuspe x x d
(Sprague)
Roberty
Rudgea ciliata (Ruiz & x x x X d
Pav.) Spreng.
Sacoglottis gabonensis x o 1 1 d
(Baill.) Urb.
Sarcocephalus pobeguinii Hua x d
ex Pobég
Sarcophrynium prionogonium o o o x d
(K.Schum.)
K.Schum.
Scottellia coriacea x d
A.Chev. & al.
Scytopetalum tieghemii x d
Hutch. &
Dalziel
Strombosia glaucescens x d
Engl.
Strychnos aculeata Soler. x x X x x d

Syzygium guineensis 3 fh
(Willd.) DC.
Syzygium paniculatum x 2 2 x 1 1 fh
Gaertn.
Tamitia utilis X d

Treculia africana Decne. x x X 2 x d

ex Trécul
Trichophyton Sp. x d

Trichoscypha arborea x 3 d
(A.Chev.)
A.Chev.
Triclisia macrophylla x 1 d
(Baill.) Diels
Tristemma hirtum x d
P.Beauv.
Uapaca corbisieri x x X x d
DeWild.
Uapaca guineensis x fn
Müll.Arg.
Uvariastrum pierreanum x x 1 d
Engl. & Diels
Vitex doniana Sweet x x fn

Xylia evansii Hutch. x o 1 d

Xylopia quintas x x d

Xylopia villosa Chipp x d

Zanha golungensis x x d
Hiern
Fungus
Agaricus bispourus x d
(J.E.Lange)
Auricularia auricula-judae. x x d
(Bull.)
J.Schröt.

180
Appendix table 5: Additional details of Chimpanzee calculus samples. Recovered plant microremains, both in
the full sample and per milligram of calculus with cause of death of the sampled chimpanzees, colour and
condition of their dental calculus and skeleton treatment during curation. Cur: Curation a) Buried for
unknown duration, cleaned and dried (1984-1994, 1996-2004 b) Necropsy, burial for 1 year, possible boiling
and dried (1994-1996) and c) Necropsy, burial for 1 year, disinfection with chlorine, 10 % formalin and dried
(2004- onwards).

Name Phytolith Starch Unsilicified Cause of death Colour Cur

Remains
Total /mg To /mg Total /mg
tal
Ophelia 0 0 1 40 0 0 Pneumonia White C
Leonardo 0 0 0 0 0 0 Starvation White/grey A
Bambou 0 0 0 0 1 7.41 Tree fall White A
Piment 0 0 0 0 0 0 Ebola White B
Oreste 40 74.63 4 7.46 1 1.87 Pneumonia Grey C
Hector 24 34.83 2 2.9 6 8.71 Anthrax Orange A
Noah 47 52.51 2 2.23 32 35.75 Unknown Brownish A
Lefkas 19 31.93 11 18.49 13 21.85 Pneumonia White A
Tina 29 21.21 8 5.85 6 4.39 Leopard Brownish A
Dorry 159 214.29 5 6.74 4 5.39 Unknown White A
Zerlina 147 167.43 0 0 9 10.25 Ebola? Moderate B
Clyde 27 23.87 4 3.54 3 2.65 Poacher White A
Agathe 94 15.47 13 2.14 22 3.62 Ebola? Brown/creamy A
Bijou 87 17.26 10 1.98 22 4.36 Unknown disease Brownish A
Leo 126 116.13 5 4.61 9 8.29 Unknown Brownish A
Castor 65 9.31 25 3.58 6 0.86 Pneumonia White A
Fanny 109 27.84 54 13.79 11 2.81 Ebola? White brown B
Kendo 233 235.59 0 0 25 25.28 Ebola? Grey B
Venus 96 59.26 16 9.88 2 1.23 Unknown Brownish C
Goma 98 7.42 18 13.7 17 1.29 Anthrax White A
1
Rubra 120 17.78 10 1.48 30 4.44 Anthrax? Mixed/white C
Ondine 26 17 0 0 10 6.54 Ebola? Brown/ green A
Mkubwa 11 33.95 0 0 1 3.09 Unknown Whitish green A
Brutus 161 49.6 5 1.54 25 7.7 Unknown Brownish A

181
Appendix table 6: Metrics of reference phytoliths and starches. Phytoliths=first part of table. Starches=second
part of table.
Species

Length

Width

LW Ratio

Brea

Area

Irregular

Spinelen

Spineno

Spineang

Shape

Conjoined
Elaeis 13.0 10.9 1.2 10.9 108.3 1 0.96 24 81 prolate 1

Elaeis 8.5 6.1 1.4 6.1 35.7 1 0.76 13 80 ovoid 1

Elaeis 10.4 9.2 1.1 9.2 71.4 2 1.2 17 98 prolate 1

Elaeis 9.3 8.6 1.1 8.6 50.8 3 0.9 14 75 spherical 1

Elaeis 12.5 10.3 1.2 10.3 105.8 2 0.95 16 80 spherical 1

Elaeis 13.3 10.5 1.3 10.5 115.5 2 1.2 19 78 prolate 1

Elaeis 8.4 7.7 1.1 7.7 45.8 1 0.68 18 83 spherical 1

Elaeis 12.7 10.0 1.3 10.0 100.0 4 1 16 95 spherical 1

Elaeis 17.8 16.7 1.1 16.7 246.0 3 1.74 18 96.25 spherical 1

Elaeis 15.5 15.0 1.0 15.0 210.1 2 2 13 94 spherical 1

Elaeis 8.6 8.6 1.0 8.6 59.0 3 1.05 14 85 spherical 1

Elaeis 11.7 8.1 1.4 8.1 75.7 4 1.2 20 90 ovoid 1

Elaeis 10.8 7.9 1.4 7.9 70.2 4 1.02 14 80.59 ovoid 1

elaeis 12.1 11.2 1.1 11.2 125.6 4 1.33 24 83 spherical 1

elaeis 7.3 6.5 1.1 6.5 46.0 4 1.13 11 103 ovoid 1

elaeis 11.0 8.3 1.3 8.3 84.6 3 1.23 17 84.43 prolate 1

elaeis 13.2 11.5 1.1 11.5 107.4 2 1.74 11 103 prolate 1

elaeis 7.8 7.2 1.1 7.2 47.1 4 1.74 10 64 spherical 1

elaeis 6.5 5.6 1.1 5.6 29.8 8 0.63 10 85 ovoid 1

elaeis 11.2 8.4 1.3 8.4 82.0 3 1.19 13 80.86 prolate 1

elaeis 13.2 11.4 1.2 11.4 125.4 3 1.17 11 82 prolate 1

elaeis 11.2 9.5 1.2 9.5 87.8 4 1.23 20 83.97 prolate 1

elaeis 8.6 7.1 1.2 7.1 45.6 5 0.92 13 99.37 ovoid 1

elaeis 9.3 6.9 1.4 6.9 49.5 4 0.79 12 69.05 prolate 1

elaeis 10.6 9.0 1.2 9.0 75.6 3 1.3 14 75.56 spherical 1

elaeis 7.8 6.2 1.2 6.2 33.0 4 0.89 14 102 ovoid 1

elaeis 5.4 5.3 1.0 5.3 22.5 4 1.1 6 96 polygon 1

elaeis 7.8 5.9 1.3 5.9 33.4 2 1.03 14 85 ovoid 1

elaeis 7.1 4.8 1.5 4.8 20.0 4 0.8 11 93 polygon 1

elaeis 7.3 3.5 2.1 3.5 17.2 4 0.84 3 50 elongate 1

elaeis 3.3 2.2 1.5 2.2 5.4 4 0.37 3 94 polygon 1

elaeis 5.7 4.1 1.4 4.1 18.4 5 0 0 0 polygon 1

elaeis 7.5 5.6 1.3 5.6 29.6 3 0.94 8 84.36 polygon 1

elaeis 6.7 4.2 1.6 4.2 20.7 4 0.6 10 70 polygon 1

elaeis 6.3 4.5 1.4 4.5 15.5 4 0.9 12 93 elongate 1

elaeis 6.5 5.8 1.1 5.8 34.4 2 0.89 12 110 polygon 1

elaeis 5.5 3.1 1.8 3.1 12.6 2 0.77 14 80 elongate 1

elaeis 7.4 3.1 2.4 3.1 16.4 3 0.78 7 81 elongate 1

elaeis 11.0 8.6 1.3 8.6 66.6 4 1.28 8 71 polygon 1

elaeis 7.4 4.1 1.8 4.1 21.3 2 0.63 10 95 prolate 1

182
elaeis 7.4 5.0 1.5 5.0 24.9 4 0.92 10 86 polygon 1

elaeis 9.4 8.0 1.2 8.0 54.6 1 1.1 16 108 prolate 1

elaeis 8.0 6.1 1.3 6.1 34.2 3 0.6 13 115 prolate 1

elaeis 7.5 5.5 1.4 5.5 30.7 3 0.72 13 98 ovoid 1

elaeis 9.1 6.3 1.4 6.3 41.8 4 0.94 12 75.83 elongate 1

elaeis 5.7 4.0 1.4 4.0 16.0 5 0.87 11 77 polygon 1

elaeis 8.3 6.4 1.3 6.4 36.4 4 0.7 11 102 polygon 1

elaeis 8.7 7.1 1.2 7.1 54.1 4 0.87 19 101 ovoid 1

elaeis 8.3 7.3 1.1 7.3 51.0 4 0.62 9 101 polygon 1

elaeis 6.8 6.2 1.1 6.2 33.0 4 0.92 15 82.61 ovoid 1

eremo 7.8 7.6 1.0 7.6 41.4 1 1 9 70 spherical 1

eremo 6.9 6.5 1.1 6.5 36.2 2 0.7 10 115 spherical 1

eremo 6.3 5.6 1.1 5.6 25.1 3 0.7 6 96 prolate 1

eremo 7.5 7.2 1.0 7.2 37.2 1 0.87 10 90 spherical 1

eremo 8.2 6.1 1.3 6.1 40.7 2 0.78 8 82 ovoid 1

eremo 7.5 7.3 1.0 7.3 42.6 1 0.87 9 78 prolate 1

eremo 6.2 6.1 1.0 6.1 27.5 4 0.82 5 86 spherical 1

eremo 5.5 5.1 1.1 5.1 22.4 2 0.5 6 90 spherical 1

eremo 8.1 6.3 1.3 6.3 38.1 3 0.6 9 81 prolate 1

eremo 7.0 6.0 1.2 6.0 35.0 2 0.94 10 70 prolate 1

eremo 7.8 7.6 1.0 7.6 40.1 2 1.7 8 90 spherical 1

eremo 5.9 4.4 1.3 4.4 23.1 4 0.88 6 89 ovoid 1

eremo 6.7 6.6 1.0 6.6 31.0 2 0.92 9 79 spherical 1

eremo 5.4 4.3 1.3 4.3 17.9 2 0.5 2 112 prolate 1

eremo 6.6 6.2 1.1 6.2 25.0 2 0.64 7 110 spherical 1

eremo 6.6 5.4 1.2 5.4 27.0 2 0.68 12 0.96 spherical 1

eremo 6.2 5.4 1.1 5.4 24.2 1 0.79 8 109 spherical 1

eremo 6.2 4.9 1.3 4.9 21.3 2 0.63 8 99 triangular 1

eremo 4.9 4.1 1.2 4.1 12.6 5 0.51 4 104 triangular 1

eremo 7.6 7.1 1.1 7.1 33.8 1 1.6 8 80 spherical 1

eremo 7.1 6.3 1.1 6.3 36.0 2 0.88 9 99 spherical 1

eremo 4.5 4.2 1.0 4.2 16.3 1 0.41 7 110 spherical 1

eremo 7.4 6.1 1.2 6.1 34.3 3 0.68 10 105 polygon 1

eremo 5.5 5.2 1.1 5.2 22.8 2 0.61 5 86 spherical 1

eremo 7.5 7.5 1.0 7.5 38.2 2 0.94 8 62.81 spherical 1

eremo 5.9 5.6 1.1 5.6 27.6 2 1.02 8 91 spherical 1

eremo 6.5 4.8 1.3 4.8 21.1 4 0.83 9 114 ovoid 1

eremo 5.6 5.1 1.1 5.1 22.1 3 0.72 3 111 spherical 1

eremo 6.0 5.0 1.2 5.0 25.6 4 0.94 9 91 spherical 1

eremo 5.9 5.9 1.0 5.9 26.2 3 0.92 9 99 spherical 1

eremo 3.7 3.6 1.0 3.6 12.6 3 0.61 4 127 spherical 1

eremo 6.2 6.2 1.0 6.2 34.7 3 1.02 12 99.69 spherical 1

eremo 6.5 4.8 1.4 4.8 28.1 3 0.83 8 95 prolate 1

eremo 7.9 6.7 1.2 6.7 44.2 3 0.94 13 84.53 spherical 1

eremo 6.1 4.0 1.5 4.0 19.2 3 0.69 8 122.88 prolate 1

eremo 6.8 6.1 1.1 6.1 33.5 4 1.02 6 110 polygon 1

183
eremo 5.6 5.1 1.1 5.1 21.9 3 0.74 6 92 spherical 1

eremo 4.9 4.4 1.1 4.4 17.5 2 0.5 8 89 spherical 1

eremo 4.6 4.4 1.0 4.4 20.2 5 0.83 5 95 polygon 1

eremo 7.9 6.2 1.3 6.2 35.6 4 0.83 10 102 polygon 1

eremo 5.5 5.2 1.1 5.2 22.1 3 0.52 7 124 spherical 1

eremo 6.1 5.5 1.1 5.5 30.9 4 0.95 7 92 polygon 1

eremo 3.9 3.3 1.2 3.3 10.1 3 0.74 2 94 prolate 1

eremo 6.9 6.7 1.0 6.7 38.8 4 1.17 10 90.68 spherical 1

eremo 5.8 4.7 1.2 4.7 26.1 4 0.83 6 75 polygon 1

eremo 3.9 3.3 1.2 3.3 10.6 5 0.62 7 85 polygon 1

eremo 4.5 3.4 1.3 3.4 13.7 4 0.66 2 107 polygon 1

eremo 6.2 5.3 1.2 5.3 27.1 3 0.66 8 87.47 prolate 1

eremo 6.5 6.1 1.1 6.1 36.2 3 0.8 11 36.22 spherical 1

eremo 6.5 4.1 1.6 4.1 24.4 5 0.72 6 91.39 polygon 1

aframomum 10.9 9.9 1.1 9.9 85.0 3 0.6 6 0 ovoid 1

aframomum 11.4 10.2 1.1 10.2 86.6 3 0 0 0 spherical 1

aframomum 10.5 7.6 1.4 7.6 66.5 4 0.55 5 0 ovoid 1

aframomum 8.4 7.5 1.1 7.5 70.7 4 0 0 0 quadrangular 1

aframomum 9.1 6.0 1.5 6.0 50.0 3 0 0 0 prolate 1

aframomum 10.0 6.1 1.6 6.1 48.6 3 0 0 0 prolate concave-convex 1

aframomum 11.4 10.4 1.1 10.4 94.5 4 0 0 0 spherical 1

aframomum 10.1 9.5 1.1 9.5 83.7 3 0 0 0 spherical 1

aframomum 14.2 9.7 1.5 9.7 14.2 4 0.5 10 0 ovoid 1

aframomum 9.9 6.6 1.5 6.6 52.5 3 0 0 0 quadrangular 1

aframomum 10.4 6.1 1.7 6.1 53.9 3 0 0 0 quadrangular 1

aframomum 11.9 10.2 1.2 10.2 96.3 3 0.55 14 0 spherical 2

aframomum 12.5 7.4 1.7 7.4 82.7 4 0.6 11 0 prolate 1

aframomum 8.5 5.5 1.5 5.5 47.8 4 0 0 0 polygon 2

aframomum 9.5 7.3 1.3 7.3 62.0 4 0 0 0 prolate concave-convex 1

aframomum 10.5 5.2 2.0 5.2 53.4 5 0 0 0 prolate concave-convex 1

aframomum 7.2 3.5 2.1 3.5 28.2 2 0 0 0 polygon concave 1

aframomum 9.6 6.3 1.5 6.3 43.1 4 0.75 9 0 prolate 1

aframomum 8.8 6.6 1.3 6.6 54.4 0 0 0 0 polygon concave 1

aframomum 7.4 4.7 1.6 4.7 31.0 0 0 0 0 polygon 1

aframomum 8.1 6.3 1.3 6.3 39.7 0 0 0 0 polygon 1

aframomum 11.0 9.6 1.1 9.6 83.5 4 0.5 15 0 spherical 1

aframomum 9.1 6.6 1.4 6.6 48.2 5 0.6 5 0 prolate concave-convex 1

aframomum 9.1 7.9 1.2 7.9 55.0 5 0.7 7 0 ovoid 1

aframomum 9.0 8.1 1.1 8.1 52.4 5 0 0 0 ovoid 1

aframomum 5.7 4.9 1.1 4.9 20.7 3 0.6 3 47 spherical 1

aframomum 7.3 6.1 1.2 6.1 35.4 3 1 11 41 spherical 1

aframomum 7.2 6.3 1.1 6.3 37.3 4 0.9 8 59 spherical 1

aframomum 6.0 6.0 1.0 6.0 30.2 3 1 6 64.88 spherical 1

aframomum 7.0 5.0 1.4 5.0 22.2 4 0.4 6 99 spherical 1

aframomum 5.3 4.1 1.3 4.1 20.4 4 0.4 3 47 spherical 1

aframomum 5.2 5.1 1.0 5.1 22.6 4 0.4 2 41 quadrangular 1

184
aframomum 5.5 4.7 1.2 4.7 20.7 1 0.26 1 54 spherical 1

aframomum 6.0 5.7 1.0 5.7 26.2 3 0.5 5 89 spherical 1

aframomum 5.8 4.6 1.3 4.6 23.5 4 0 0 0 spherical 1

aframomum 5.6 4.2 1.3 4.2 12.9 3 0.6 4 47 spherical 1

aframomum 5.5 5.0 1.1 5.0 28.8 4 0.72 8 64 spherical 1

aframomum 9.4 7.1 1.3 7.1 44.4 3 0 0 0 angularpoint 1

aframomum 6.0 5.1 1.2 5.1 20.2 0 0 0 0 spherical 1

aframomum 6.9 4.4 1.6 4.4 26.2 5 0 0 0 spherical 1

aframomum 6.4 4.3 1.5 4.3 21.6 4 0.55 6 66 spherical 1

aframomum 6.0 5.6 1.1 5.6 28.3 3 0.6 6 82 spherical 1

aframomum 5.9 5.4 1.1 5.4 28.0 3 0.4 7 70 spherical 1

aframomum 6.1 3.5 1.7 3.5 34.5 4 0.7 10 53 spherical 1

aframomum 6.8 6.0 1.1 6.0 33.0 3 0.7 8 82 spherical 1

aframomum 6.6 5.8 1.1 5.8 30.0 3 0.52 7 77 spherical 1

aframomum 5.8 4.9 1.2 4.9 22.2 5 0.83 5 41 spherical 1

aframomum 6.4 5.7 1.1 5.7 22.1 5 0.83 4 42 polygon 1

aframomum 6.5 5.5 1.2 5.5 30.2 3 0.9 8 78 spherical 1

aframomum 7.2 6.3 1.1 6.3 40.0 4 0.5 6 100 ovoid 1

ancistrophy 6.0 6.0 1.0 6.0 27.8 2 0.7 7 116 spherical 1

ancistrophy 5.4 4.2 1.3 4.2 18.9 4 0.55 4 105 ovoid 1

ancistrophy 4.9 4.8 1.0 4.8 21.2 2 0.58 4 116 spherical 1

ancistrophy 3.7 3.2 1.2 3.2 10.2 2 0 0 0 polygon 1

ancistrophy 5.9 4.4 1.3 4.4 19.8 4 0.46 3 191 ovoid 1

ancistrophy 3.2 2.5 1.3 2.5 7.2 4 0 0 0 polygon 1

ancistrophy 4.8 4.5 1.1 4.5 18.0 3 0.51 3 90 polygon 1

ancistrophy 5.8 5.7 1.0 5.7 22.9 4 0.75 4 108.06 polygon 1

ancistrophy 5.0 4.1 1.2 4.1 15.4 3 0.52 3 120.35 polygon 1

ancistrophy 4.9 2.9 1.7 2.9 12.4 5 0 0 0 elongate 1

ancistrophy 3.5 2.8 1.3 2.8 8.2 5 0 0 0 polygon 1

ancistrophy 3.2 2.3 1.4 2.3 5.3 5 0 0 0 polygon 1

ancistrophy 5.8 5.0 1.2 5.0 18.8 4 0.51 7 0 spherical 1

ancistrophy 4.9 4.2 1.2 4.2 16.8 3 0.32 3 95.28 prolate 1

ancistrophy 4.1 3.3 1.2 3.3 11.7 5 0.5 4 91 polygon 1

ancistrophy 4.7 3.7 1.3 3.7 13.3 4 0 0 0 prolate 1

ancistrophy 4.5 3.6 1.2 3.6 11.5 3 0.4 3 118 polygon 1

ancistrophy 5.2 3.5 1.5 3.5 16.3 5 0.62 4 87 ovoid 1

ancistrophy 3.6 2.8 1.3 2.8 8.1 4 0 0 0 ovoid 1

ancistrophy 3.4 2.8 1.2 2.8 7.5 4 0 0 0 spherical 1

ancistrophy 5.1 3.9 1.3 3.9 15.3 4 0.55 3 109 polygon 1

ancistrophy 4.3 3.9 1.1 3.9 13.4 3 0.4 3 116 spherical 1

ancistrophy 3.6 2.9 1.3 2.9 8.0 5 0 0 0 polygon 1

ancistrophy 3.8 2.9 1.3 2.9 9.6 5 0 0 0 polygon 1

ancistrophy 4.9 3.8 1.3 3.8 13.5 5 0.3 4 111 polygon 1

ancistrophy 7.2 4.2 1.7 4.2 27.3 4 0.62 9 27.27 prolate 1

ancistrophy 5.6 5.3 1.0 5.3 24.0 2 1.07 5 61.83 spherical 1

ancistrophy 7.3 6.5 1.1 6.5 43.5 1 1.2 8 83.46 spherical 1

185
ancistrophy 8.9 7.6 1.2 7.6 53.5 1 1.1 12 85.57 spherical 1

ancistrophy 6.1 5.8 1.0 5.8 29.5 3 1.27 8 82 spherical 1

ancistrophy 5.3 4.3 1.2 4.3 20.2 4 1.25 8 76.85 spherical 1

ancistrophy 7.2 5.2 1.4 5.2 34.2 2 1.05 9 79.57 prolate 1

ancistrophy 7.5 5.0 1.5 5.0 29.1 4 0.88 6 73.5 ovoid 1

ancistrophy 8.5 7.2 1.2 7.2 44.5 1 1.11 10 86.09 spherical 1

ancistrophy 5.8 5.4 1.1 5.4 24.9 3 0.88 9 106 spherical 1

ancistrophy 7.5 5.9 1.3 5.9 38.1 3 1.23 9 85 spherical 1

ancistrophy 7.5 6.5 1.2 6.5 36.8 3 1.2 7 81 spherical 1

ancistrophy 6.3 5.8 1.1 5.8 32.5 3 1.11 7 84.33 spherical 1

ancistrophy 5.9 5.4 1.1 5.4 27.4 3 0.97 9 69.85 ovoid 1

ancistrophy 7.0 4.9 1.4 4.9 31.0 5 0.65 7 96 spherical 1

ancistrophy 9.7 8.8 1.1 8.8 82.7 2 1.64 11 92 spherical 1

ancistrophy 7.6 7.1 1.1 7.1 47.5 2 1.25 9 86.23 spherical 1

ancistrophy 8.8 6.2 1.4 6.2 49.3 3 1.33 10 82.53 prolate 1

ancistrophy 6.5 6.2 1.0 6.2 35.8 3 1.02 8 84.17 spherical 1

ancistrophy 7.0 6.6 1.1 6.6 35.3 4 1.24 6 88.91 spherical 1

ancistrophy 6.8 5.8 1.2 5.8 30.9 3 0.97 6 97 spherical 1

ancistrophy 5.6 5.6 1.0 5.6 24.5 3 0.55 7 115 spherical 1

ancistrophy 5.4 3.8 1.4 3.8 17.9 4 0.92 5 77 prolate 1

ancistrophy 8.9 6.9 1.3 6.9 57.3 3 1.2 2 72 prolate 1

ancistrophy 7.9 5.8 1.4 5.8 37.5 4 1.5 10 84 spherical 1

sarcoph 17.5 6.6 2.7 6.6 87.8 4 0 0 0 angularpoint 1

sarcoph 15.5 6.1 2.6 6.1 63.0 3 0 0 0 angularpoint 1

sarcoph 16.5 6.3 2.6 6.3 81.6 3 0 0 0 angularpoint 1

sarcoph 16.3 7.6 2.2 7.6 82.0 3 0 0 0 angularpoint 1

sarcoph 14.4 7.2 2.0 7.2 55.6 4 0 0 0 angularpoint 1

sarcoph 14.7 7.4 2.0 7.4 73.3 5 0 0 0 angularelongate 1

sarcoph 19.8 7.5 2.6 7.5 108.6 5 0 0 0 angularpoint 1

sarcoph 19.4 7.2 2.7 7.2 70.0 4 0 0 0 angularpoint 1

sarcoph 19.3 7.3 2.6 7.3 92.5 4 0 0 0 angularpoint 1

sarcoph 14.3 5.3 2.7 5.3 66.5 3 0 0 0 angularpoint 1

sarcoph 14.2 6.7 2.1 6.7 83.5 3 0 0 0 quadrangular 1

sarcoph 14.8 12.9 1.1 12.9 133.9 5 0 0 0 triangular 1

sarcoph 15.9 5.7 2.8 5.7 76.7 4 0 0 0 angularpoint 1

sarcoph 14.1 8.7 1.6 8.7 89.1 4 0 0 0 angularpoint 1

sarcoph 16.1 6.2 2.6 6.2 87.1 4 1.68 1 40 angularpoint 1

sarcoph 16.7 6.4 2.6 6.4 92.0 4 0 0 0 angularpoint 1

sarcoph 15.0 7.8 1.9 7.8 82.2 4 0 0 0 angularpoint 1

sarcoph 11.8 7.6 1.6 7.6 57.4 5 0 0 0 angularpoint 1

sarcoph 16.2 6.3 2.6 6.3 80.8 5 0 0 0 angularelongate 1

sarcoph 11.6 5.1 2.3 5.1 56.8 4 0 0 0 angularpoint 1

sarcoph 10.3 7.6 1.3 7.6 49.5 4 0 0 0 angularpoint 1

sarcoph 16.4 6.6 2.5 6.6 95.1 4 0 0 0 angularelongate 1

sarcoph 19.3 6.3 3.1 6.3 74.7 4 0 0 0 angularelongate 1

sarcoph 20.4 7.6 2.7 7.6 136.4 4 0 0 0 angularelongate 1

186
sarcoph 14.4 6.8 2.1 6.8 81.0 3 0 0 0 angularelongate 1

sarcoph 14.7 8.6 1.7 8.6 87.8 3 0 0 0 angularelongate 1

sarcoph 14.7 8.6 1.7 8.6 71.7 4 0 0 0 angularelongate 1

sarcoph 23.6 21.7 1.1 21.7 294.8 3 5 12 40.2 spherical 1

sarcoph 14.0 6.9 2.0 6.9 89.5 3 0 0 0 quadrangular 1

sarcoph 21.3 8.8 2.4 8.8 130.9 4 0 0 0 angularpoint 1

sarcoph 13.3 7.2 1.8 7.2 75.9 5 0 0 0 angularelongate 1

sarcoph 19.5 8.4 2.3 8.4 114.5 5 0 0 0 angularpoint 1

sarcoph 11.5 4.6 2.5 4.6 37.9 3 0 0 0 angularpoint 1

sarcoph 13.8 7.6 1.8 7.6 71.4 4 0 0 0 angularpoint 1

sarcoph 15.2 6.5 2.3 6.5 84.9 4 0 0 0 quadrangular 1

sarcoph 15.3 9.3 1.6 9.3 127.6 3 0 0 0 quadrangular 1

sarcoph 11.5 6.6 1.8 6.6 51.2 4 0 0 0 angularpoint 1

sarcoph 13.8 6.2 2.2 6.2 56.3 4 0 0 0 angularpoint 1

sarcoph 11.4 6.6 1.7 6.6 56.1 3 0 0 0 quadrangular 1

sarcoph 16.8 6.0 2.8 6.0 78.3 4 0 0 0 angularpoint 1

sarcoph 19.7 6.2 3.2 6.2 94.7 4 0 0 0 angularpoint 1

sarcoph 13.8 6.7 2.0 6.7 74.2 5 0 0 0 angularpoint 1

sarcoph 17.6 9.4 1.9 9.4 94.9 4 0 0 0 angularpoint 1

sarcoph 13.8 8.4 1.6 8.4 105.4 3 0 0 0 quadrangular 1

sarcoph 17.3 6.6 2.6 6.6 70.2 4 0 0 0 angularpoint 1

sarcoph 16.6 9.3 1.8 9.3 91.9 4 2.3 1 58 angularpoint 1

sarcoph 15.2 7.5 2.0 7.5 89.7 3 0 0 0 angularpoint 1

sarcoph 16.5 7.7 2.1 7.7 101.4 5 0 0 0 angularpoint 1

sarcoph 20.2 7.8 2.6 7.8 118.8 5 0 0 0 angularpoint 1

sarcoph 12.2 6.5 1.9 6.5 57.2 5 0 0 0 angularpoint 1

Metrics of reference starches.

Species

Length

Width

LW Ratio

Brea

Area

Shape

Facets

Striaelen

Striaeno

Type

Lam

Dist

cola 27.82 23.59 1.18 18.9 515 ovoid 0 6.48 1 1 3 12.19

cola 21.85 20.68 1.06 13 396.24 elongate 0 5.86 2 2 3 10.24

conovoid
cola 13.13 8.4 1.56 8.4 70.65 pyriform 0 0 0 1 2 6.56

cola 11.87 8.54 1.39 8.54 82.8 elongate 0 0 0 2 2 6.04

conovoid
cola 11.16 8.61 1.3 8.61 75.1 elongate 0 0 0 2 2 4.72
conovoid
cola 3.42 3.42 1 2.39 7.52 oblate 0 0 0 2 0 1.13
conovoid
cola 22.39 17.83 1.26 12.3 300.88 ovoid 0 4.36 1 1 2 8.33

cola 9.19 7.6 1.21 7.6 61.36 prolate 1 0 0 1 0 3.25

cola 8.41 6.67 1.26 6.67 42.84 prolate 1 0 0 1 0 2.59

cola 11.68 8.94 1.31 7.65 79.98 triangular 0 0 0 1 0 4.65

cola 12.73 9.8 1.3 9.8 88.7 prolate 1 2.62 1 1 0 5.02

cola 5.86 4.51 1.3 4.51 20.95 oblate 2 0 0 2 0 2.06

conovoid
cola 13.18 9.99 1.32 9.99 105.93 ovoid 0 0 0 1 0 5.88

187
cola 6.04 4.92 1.23 4.92 25.57 ovoid 1 0 0 1 0 2.62

cola 4.18 3.21 1.3 3.21 15.26 ovoid 1 0 0 1 0 1.23

cola 7.62 7.62 1 6.69 103.45 oblate 2 0 0 2 0 4.09

conovoid
cola 4.75 3.69 1.29 3.69 21.38 prolate 1 0 0 1 0 1.79

cola 3.57 3.57 1 2.7 9.79 oblate 1 0 0 2 0 1.35

conovoid
cola 11.59 9.34 1.24 9.34 83.96 ovoid 0 0 0 1 0 4.61

cola 8.78 7.92 1.11 7.92 58.53 ovoid 0 0 0 1 0 3.71

cola 4.85 4.08 1.19 4.08 18.63 polygon 3 0 0 2 0 1.33

cola 6.4 4.42 1.45 4.42 22.9 ovoid 2 0 0 1 0 1.85

cola 27.23 19.54 1.39 18 402 ovoid 0 8 4 1 2 11.84

cola 26.46 17.45 1.52 17.5 346.76 ovoid 0 7 5 1 0 15.59

cola 14.34 12.99 1.1 12.99 143.36 prolate 0 0 0 1 2 6.18

cola 20.34 14.36 1.42 14.4 226.29 ovoid 0 7.8 4 1 0 9.62

cola 26.75 19.64 1.36 19.64 408.6 ovoid 0 9.87 3 1 3 15.02

cola 14.7 11.69 1.26 11.69 144.5 elongate 0 2.27 1 2 2 7.62

conovoid
cola 10.22 8.71 1.17 8.71 68.55 prolate 0 0 0 1 0 2.91

cola 9.52 7.2 1.32 7.2 64.07 ovoid 0 0 0 1 2 2.97

cola 8.26 6.45 1.28 6.45 42.47 ovoid 0 0 0 1 0 2.46

cola 10.59 7.91 1.34 7.91 76.69 ovoid 0 0 0 1 2 4.12

cola 11.43 8.14 1.4 8.14 73.02 pyriform 0 0 0 1 0 4.16

cola 8.91 6.6 1.35 6.6 48.02 ovoid 1 0 0 1 0 2.69

cola 7.45 5.73 1.3 5.73 39.98 ovoid 0 0 0 1 0 2.67

cola 5.6 3.7 1.51 3.7 18.25 prolate 0 0 0 1 0 1.37

cola 6.52 4.86 1.34 4.86 45 polygon 3 0 0 2 0 1.83

cola 14.55 14.55 1 10.4 119.07 elongate 0 2.9 2 2 2 7.89

conovoid
cola 8.7 6.15 1.41 6.15 39.45 ovoid 0 0 0 1 0 3.8

cola 12.24 9.38 1.3 9.38 89.29 ovoid 0 1.62 2 1 0 1.85

cola 12.82 10.55 1.22 10.55 120.54 prolate 0 0 0 1 0 4.51

cola 6.6 5.03 1.31 5.03 25.85 ovoid 0 0 0 1 0 2.46

cola 6.65 5.36 1.24 5.36 50.82 triangular 0 0 0 1 0 1.84

cola 15.39 9.32 1.65 9.32 99.78 pyriform 0 0 0 1 0 3.69

cola 4.5 3.69 1.22 3.69 16.32 spherical 0 0 0 1 0 2.15

cola 14.47 12.81 1.13 12.81 130.47 prolate 0 0 0 1 0 3.07

cola 3.6 3.08 1.17 3.08 11.8 spherical 0 0 0 1 0 1.45

cola 7.32 5.32 1.38 5.32 33.19 oblate 2 0 0 1 0 1.65

conovoid
cola 13.82 10.61 1.3 10.61 104.83 ovoid 0 0 0 1 0 2.97

cola 16.98 11.45 1.48 11.45 128.62 triangular 0 0 0 1 0 2.46

aframomum 2.86 2.73 1.05 2.73 4.95 hemispherical 1 0 0 3 0 1.43

aframomum 2.36 2.1 1.12 2.1 3.87 spherical 0 0 0 3 0 1.18

aframomum 2.39 2.18 1.1 2.18 3.71 spherical 1 0 0 3 0 1.195

aframomum 2.32 1.99 1.17 1.99 4.08 spherical 0 0 0 3 0 1.16

aframomum 3.13 2.83 1.11 2.83 4.7 spherical 0 0 0 3 0 1.565

aframomum 2.47 2.18 1.13 2.18 4.1 spherical 0 0 0 3 0 1.235

aframomum 3.26 2.85 1.14 2.85 7.5 spherical 1 0 0 3 0 1.63

aframomum 2.88 2.57 1.12 2.57 5.59 spherical 1 0 0 3 0 1.44

188
aframomum 4.81 4.45 1.08 4.45 16 spherical 2 0 0 3 0 2.405

aframomum 3.52 3.4 1.04 3.4 7.18 spherical 2 0 0 3 0 1.76

aframomum 2.57 2.46 1.04 2.46 4.36 spherical 1 0 0 3 0 1.285

aframomum 2.46 2.15 1.14 2.15 3.5 spherical 1 0 0 3 0 1.23

aframomum 3.25 2.4 1.35 2.4 5.67 ovoid 0 0 0 3 0 1.625

aframomum 2.9 2.1 1.38 2.1 4.91 spherical 1 0 0 3 0 1.45

aframomum 2.54 2.21 1.15 2.21 4.31 spherical 2 0 0 3 0 1.27

aframomum 2.73 2.63 1.04 2.63 5.54 spherical 2 0 0 3 0 1.365

aframomum 2.5 2.2 1.14 2.2 5.99 spherical 3 0 0 3 0 1.25

aframomum 2.15 2.02 1.06 2.02 3.8 spherical 2 0 0 3 0 1.075

aframomum 2.95 2.16 1.37 2.16 5.55 spherical 1 0 0 3 0 1.475

aframomum 3.71 3.56 1.04 3.56 10 spherical 2 0 0 3 0 1.855

aframomum 3.49 2.98 1.17 2.98 6.23 spherical 2 0 0 3 0 1.745

aframomum 3.69 3.69 1 3.69 9.25 spherical 1 0 0 3 0 1.845

aframomum 3.49 3.19 1.09 3.19 9.97 spherical 2 0 0 3 0 1.745

aframomum 1.99 1.87 1.06 1.87 2.71 spherical 1 0 0 3 0 0.995

aframomum 3.04 2.58 1.18 2.58 5.08 spherical 0 0 0 3 0 1.52

aframomum 5.65 5.45 1.04 5.45 21.99 spherical 2 0 0 3 0 2.825

aframomum 3.85 2.82 1.37 2.82 8.69 spherical 2 0 0 3 0 1.925

aframomum 2.66 2.05 1.3 2.05 4.82 hemispherical 1 0 0 3 0 1.33

aframomum 2.86 2.21 1.29 2.21 4.96 hemispherical 2 0 0 3 0 1.43

aframomum 2.83 2.53 1.12 2.53 6.78 spherical 2 0 0 3 0 1.415

aframomum 3.11 2.88 1.08 2.88 9.42 spherical 3 0 0 3 0 1.555

aframomum 1.9 1.68 1.13 1.68 3 spherical 3 0 0 3 0 0.95

aframomum 3.58 3.38 1.06 3.38 9.45 spherical 1 0 0 3 0 1.79

aframomum 1.78 1.54 1.16 1.54 1.8 spherical 1 0 0 3 0 0.89

aframomum 2.05 1.64 1.25 1.64 2.22 spherical 2 0 0 3 0 1.025

aframomum 2.78 2.36 1.18 2.36 3.99 spherical 2 0 0 3 0 1.39

aframomum 1.81 1.69 1.07 1.69 2.57 spherical 0 0 0 3 0 0.905

aframomum 5.12 4.5 1.14 4.5 16.4 spherical 1 0 0 3 0 2.56

aframomum 2.57 1.96 1.31 1.96 3.89 spherical 1 0 0 3 0 1.285

aframomum 2.73 2.32 1.18 2.32 5.95 polygon 4 0 0 3 0 1.365

aframomum 4.27 4.08 1.05 4.08 14 polygon 4 0 0 3 0 2.135

aframomum 2.02 1.7 1.19 1.7 3.5 hemispherical 1 0 0 3 0 1.01

aframomum 2.82 2.67 1.06 2.67 6.37 spherical 1 0 0 3 0 1.41

aframomum 3.09 2.15 1.06 2.15 5.59 polygon 4 0 0 3 0 1.02

aframomum 3.49 2.87 1.06 2.87 8.31 polygon 4 0 0 3 0 1.45

aframomum 3.18 3.08 1.06 3.08 8.34 spherical 0 0 0 1 0 1.13

aframomum 4.3 3.28 1.06 3.28 11.5 spherical 1 0 0 1 0 1.13

aframomum 2.27 1.64 1.06 1.64 3.4 spherical 1 0 0 1 0 1.21

aframomum 5.03 3.18 1.06 3.18 12.83 polygon 3 0 0 3 0 1.5

aframomum 3.41 3.24 1.06 3.24 7.96 polygon 4 0 0 3 0 1.6

piper 5.15 3.8 1.36 3.8 14.67 polygon 7 0 0 3 0 2.36

piper 2.58 2.25 1.15 2.25 3.61 polygon 6 0 0 3 0 1.3

piper 4.18 2.84 1.47 2.84 9.52 polygon 5 0 0 3 0 2

piper 3.86 3.4 1.14 3.4 7.37 hemispherical 4 0 0 3 0 1.9

189
piper 3.4 2.7 1.26 2.7 7.49 polygon 7 0 0 3 0 1.7

piper 3.21 2.79 1.15 2.79 8.9 polygon 6 0 0 3 0 1.6

piper 3 2 1.5 2 10 polygon 6 0 0 3 0 1.5

piper 3.98 3.28 1.21 3.28 15 polygon 7 0 0 3 0 1.49

piper 3.95 3.58 1.1 3.58 13.99 polygon 7 0 0 3 0 1.49

piper 3.18 2.9 1.1 2.9 10.17 polygon 6 0 0 3 0 1.64

piper 3.32 3.11 1.07 3.11 9.15 polygon 6 0 0 3 0 1.67

piper 3.04 2.58 1.18 2.58 6.68 polygon 7 0 0 3 0 1.52

piper 3.25 2.27 1.43 2.27 9.73 polygon 7 0 0 3 0 1.63

piper 4.98 3.92 1.27 3.92 14.44 polygon 6 0 0 3 0 2.5

piper 4.87 4.82 1.01 4.87 17.68 hemispherical 5 0 0 3 0 2.3

piper 5.94 3.62 1.64 3.62 11.91 polygon 7 0 0 3 0 2.8

piper 4.45 3.98 1.12 3.98 16 oblate 4 0 0 3 0 2.2

conovoid
piper 4.33 3.9 1.11 3.9 12.22 oblate 4 0 0 3 0 2.15
conovoid
piper 2.84 2.17 1.31 2.17 5.07 polygon 6 0 0 3 0 1.4

piper 5.33 5.24 1.02 3 19 hemispherical 3 0 0 3 0 2.6

piper 4.66 3.92 1.19 3 16.72 hemispherical 3 0 0 3 0 2.33

piper 3.7 2.56 1.45 2.56 10.55 polygon 5 0 0 3 0 1.8

piper 2.95 2.25 1.31 2.25 8.38 oblate 3 0 0 3 0 1.8

conovoid
piper 4.23 3.11 1.36 3.11 18.74 polygon 6 0 0 3 0 2.1

piper 4.41 3.75 1.18 3.75 10.84 oblate 5 0 0 3 0 2.2

conovoid
piper 3.52 2.62 1.34 2.62 18 polygon 6 0 0 3 0 1.75

piper 4.17 3.12 1.34 3.12 11.77 polygon 6 0 0 3 0 1.07

piper 3.15 2.43 1.3 2.43 18.55 quadrangular 6 0 0 3 0 1.6

piper 3.49 2.93 1.19 2.93 10.6 polygon 7 0 0 3 0 1.7

piper 2.98 2.53 1.18 2.53 7.22 polygon 7 0 0 3 0 1.5

piper 3.71 3.41 1.09 3.41 10.94 polygon 7 0 0 3 0 1.87

piper 3.47 2.33 1.49 2.33 8.52 oblate 3 0 0 3 0 1.72

conovoid
piper 4.03 3.33 1.21 3.33 8.55 polygon 6 0 0 3 0 2

piper 5.82 5.82 1 3.62 16.57 hemispherical 2 0 0 3 0 2.8

piper 5.41 5.41 1 3.91 18.7 hemispherical 2 0 0 3 0 2.7

piper 6.01 6.01 1 5.07 18.37 oblate 3 0 0 3 0 3

conovoid
piper 3.92 3.85 1.02 3.85 11.54 polygon 7 0 0 3 0 1.95

piper 5 3.85 1.3 3.85 15.22 polygon 7 0 0 3 0 1.65

piper 3.39 3.12 1.09 3.12 13.91 quadrangular 6 0 0 3 0 2.9

piper 3.82 2.07 1.85 2.07 6.73 oblate 5 0 0 3 0 1.35

conovoid
piper 2.71 2.27 1.19 2.27 4.84 oblate 5 0 0 3 0 1.35
conovoid
piper 3.58 2.76 1.3 2.76 7.99 polygon 7 0 0 3 0 1.79

piper 2.27 2.05 1.11 2.05 6.1 polygon 6 0 0 3 0 1.13

piper 4.16 3.75 1.11 3.75 15.12 polygon 4 0 0 1 0 1.64

piper 4.23 3.85 1.1 3.85 14.17 polygon 4 0 0 3 0 2.15

piper 2.77 2.13 1.3 2.13 4.12 polygon 2 0 0 3 0 0.87

piper 4.57 3.27 1.4 3.27 15.01 polygon 4 0 0 3 0 1.85

piper 3.84 3.61 1.06 3.61 14.88 polygon 6 0 0 3 0 1.88

190
piper 3.34 3.25 1.03 3.25 7.88 polygon 5 0 0 3 0 1.54

piper 2.25 2.06 1.09 2.06 4.78 polygon 4 0 0 3 0 1.13

sacog 5.99 4.77 1.26 3 21.57 polygon 3 0 0 2 0 1.77

sacog 5.55 4.03 1.38 2 18.73 polygon 2 0 0 2 0 1.74

sacog 6.96 5.88 1.18 3.4 36.03 oblate 3 0 0 2 0 1.71

conovoid
sacog 5.03 4.18 1.2 3 16.51 quadrangular 3 0 0 1 0 1.77

sacog 5.74 5.3 1.08 5.3 32.7 polygon 3 0 0 1 0 1.77

sacog 6.06 5.69 1.07 5.69 34 polygon 3 0 0 1 0 2.46

sacog 6.93 6.02 1.15 6.02 30.27 polygon 2 0 0 1 0 2.79

sacog 6.49 6.49 1 4.57 25.49 oblate 2 0 0 2 0 2.63

conovoid
sacog 6.23 6.23 1 4.37 18.06 oblate 2 0 0 2 0 2.39
conovoid
sacog 6.81 6.81 1 3.92 19.054 oblate 2 0 0 2 0 2.15
conovoid
sacog 5.37 5.37 1 4.23 19.02 oblate 2 0 0 2 0 1.33
conovoid
sacog 4.82 4.42 1.09 3 18.94 oblate 2 0 0 2 0 1.43
conovoid
sacog 6.72 6.72 1 4.3 22.45 oblate 2 0 0 2 0 1.64
conovoid
sacog 7.41 7.25 1.02 7.25 37.49 oblate 2 0 0 2 0 2.46
conovoid
sacog 7.8 7.01 1.11 3 41.67 oblate 2 0 0 2 0 2.36
conovoid
sacog 6.51 6.51 1 5.44 24.12 oblate 2 0 0 2 0 1.96
conovoid
sacog 8.7 6.65 1.31 6.65 56.62 oblate 2 0 0 2 0 3.08
conovoid
sacog 7.75 7.7 1.01 5.21 33.52 oblate 2 0 0 2 0 2.36
conovoid
sacog 5.96 5.96 1 4.02 18.37 oblate 2 0 0 2 0 1.96
conovoid
sacog 4.92 3.82 1.29 3.82 23 quadrangular 3 0 0 1 0 1.96

sacog 3.7 3.7 1 2.9 10.5 oblate 3 0 0 2 0 1.75

conovoid
sacog 6.29 5.49 1.15 5.49 27.54 polygon 3 0 0 1 0 2.25

sacog 5.22 4.21 1.24 4.21 20.85 polygon 4 0 0 1 0 2.16

sacog 3.69 3.07 1.2 3.07 8.21 oblate 2 0 0 2 0 1.03

conovoid
sacog 4.88 4.27 1.14 4 14.7 oblate 3 0 0 2 0 1.75
conovoid
sacog 6.97 6.97 1 5.2 34.78 polygon 4 0 0 1 0 2.05

sacog 7.77 6.45 1.2 5.12 37.23 polygon 4 0 0 1 0 2.61

sacog 4.28 4.02 1.06 3.28 31.28 polygon 4 0 0 1 0 2.35

sacog 6.8 5.88 1.16 4.6 31.59 polygon 2 0 0 1 0 2.36

sacog 4.69 4.69 1 3.33 14.7 polygon 2 0 0 1 0 1.59

sacog 7.07 5.68 1.24 4.67 34.16 polygon 2 0 0 1 0 1.97

sacog 6.78 6.45 1.05 6.41 40.39 ovoid 1 0 0 1 0 2.17

sacog 6.35 6.13 1.04 5.68 24.12 polygon 3 0 0 1 0 3.01

sacog 6.74 6.26 1.08 3.98 19.55 polygon 3 0 0 1 0 2.27

sacog 6.23 4.9 1.27 4.51 19.3 polygon 2 0 0 1 0 1.81

sacog 7.63 7.63 1 4.13 27.61 hemispherical 1 0 0 2 0 2.01

sacog 7.77 7.77 1 5.15 32.7 hemispherical 1 0 0 2 0 2.05

sacog 6.79 6.1 1.11 6.1 33.09 quadrangular 1 0 0 1 0 2.05

sacog 6.99 5.79 1.21 5.79 28.54 hemispherical 2 0 0 2 0 2.15

sacog 5.37 2.88 1.86 2.88 15.25 hemispherical 1 0 0 2 0 2.97

191
sacog 4.77 3.26 1.46 3.26 14.97 hemispherical 1 0 0 2 0 2.65

sacog 5.77 3.48 1.66 3.48 14.77 hemispherical 2 0 0 2 0 2.35

sacog 5.25 4.47 1.17 4.47 21.11 hemispherical 1 0 0 2 0 1.99

sacog 5.41 4.31 1.26 4.31 17.58 hemispherical 1 0 0 2 0 1.95

sacog 5.92 5.07 1.17 5.07 22.21 hemispherical 1 0 0 2 0 2.05

sacog 5.34 5 1.07 5 18.45 polygon 5 0 0 1 0 2.59

sacog 6.04 6.04 1 6.04 27.63 spherical 1 0 0 1 0 2.14

sacog 6.18 3.95 1.56 3.95 17.37 hemispherical 1 0 0 2 0 2.98

sacog 4.66 3.16 1.47 3.16 9.24 oblate 2 0 0 2 0 2.27

conovoid
sacog 5.14 3.79 1.36 3.79 18.35 oblate 2 0 0 2 0 1.59
conovoid
panda 4.86 3.61 1.35 3.89 15.37 ovoid 0 1.96 2 1 0 2.76

panda 5.39 4.36 1.24 4.36 15.11 spherical 0 0 0 1 0 1.75

panda 6.88 5.37 1.28 5.37 31.25 ovoid 0 0 0 1 0 3.92

panda 4.6 4.6 1 4.6 16.17 ovoid 1 0 0 1 0 2.31

panda 5.32 5.02 1.06 5.02 22.8 spherical 0 0 0 1 0 2.66

panda 4.86 3.11 1.56 3.11 12.53 elongate ovoid 1 0 0 1 0 2.48

panda 6.92 5.62 1.23 5.62 37.39 ovoid 1 0 0 1 0 3.18

panda 7.68 5.05 1.52 5.05 29.8 pyriform 0 0 0 1 0 4.42

panda 7.02 5.65 1.24 5.62 29.65 prolate 0 2.69 2 1 0 3.29

panda 7.18 6.15 1.17 6.15 36.33 ovoid 0 1.6 2 1 0 3.39

panda 4.67 3.89 1.2 3.89 15.91 ovoid 0 0 0 1 0 2.61

panda 2.51 2.16 1.16 2.16 4.54 ovoid 0 0 0 1 0 1.5

panda 4.06 4.06 1 4.06 12.13 spherical 0 0 0 1 0 1.85

panda 6.49 6.49 1 6.49 33.53 spherical 0 1.01 3 1 0 2.77

panda 4.35 3.7 1.18 3.7 12.09 ovoid 0 0 0 1 0 1.3

panda 4.18 3.8 1.1 3.76 11.66 ovoid 2 0 0 1 0 1.44

panda 6.76 5.32 1.27 5.32 30.13 prolate 0 2.05 2 1 0 3.05

panda 4.52 3.69 1.22 3.69 15.1 prolate 0 1.24 2 1 0 2.06

panda 5.09 5.09 1 5.09 21.11 spherical 1 0 0 1 0 1.13

panda 3.03 2.88 1.05 2.88 8.12 ovoid 0 0 0 1 0 1.74

panda 4.94 4.23 1.17 4.23 17.6 prolate 0 0 0 1 0 1.85

panda 4.5 3.79 1.19 3.79 14.45 prolate 0 0 0 1 0 1.85

panda 7.47 6.09 1.23 6.09 39.18 prolate 0 0 0 1 0 3.55

panda 3.55 2.66 1.33 2.66 7.22 prolate 0 0 0 1 0 1.55

panda 4.1 3.69 1.11 3.69 13.07 prolate 0 0 0 1 0 1.54

panda 5.4 5.4 1 5.4 22.07 spherical 0 0 0 1 0 2.4

panda 3.17 3.17 1 3.17 8.09 spherical 0 0 0 1 0 1.33

panda 3.13 3.13 1 3.13 8.63 spherical 1 0 0 1 0 1.43

panda 5.59 4.12 1.36 4.12 19.12 prolate 0 0 0 1 0 2.76

panda 4.73 4.73 1 4.73 18.62 spherical 0 0 0 1 0 2.15

panda 4.18 3.47 1.2 3.47 12.87 prolate 1 0 0 1 0 2.29

panda 4.93 3.5 1.41 3.5 16.27 prolate 0 0 0 1 0 2.05

panda 3.18 2.67 1.19 2.67 7.8 prolate 1 0 0 1 0 1.02

panda 7.72 7.04 1.1 7.04 47.48 prolate 0 0 0 1 0 3.17

panda 4.4 3.39 1.3 3.39 15.43 prolate 0 0 0 1 0 2.09

192
panda 4.45 3.18 1.4 3.18 14.18 prolate 0 0 0 1 0 1.54

panda 4.85 4.69 1.03 4.69 18.14 spherical 0 0 0 1 0 1.65

panda 5.06 2.81 1.8 2.81 18.96 pyriform 0 0 0 1 0 1.14

panda 4 3.81 1.05 3.81 10.25 spherical 0 0 0 1 0 1.85

panda 5.02 3.17 1.58 3.17 26.04 prolate 0 0 0 1 0 2.79

panda 4.2 2.35 1.79 2.35 10.04 ovoid 0 0 0 1 0 1.54

panda 2.16 2.15 1 2.15 5 spherical 0 0 0 1 0 1.07

panda 2.87 1.84 1.56 1.84 5.7 prolate 0 0 0 1 0 1.23

panda 4.3 3.89 1.11 3.89 16.75 spherical 0 0 0 1 0 2.15

panda 2.97 2.77 1.07 2.77 7.5 spherical 0 0 0 1 0 1.33

panda 5.59 3.9 1.43 3.9 20.94 ovoid 0 2.34 1 1 0 2.11

panda 4.97 4.11 1.21 4.97 15.6 ovoid 0 0 0 1 0 2.3

panda 3.38 2.87 1.18 2.87 16.75 prolate 1 0 0 1 0 1.13

panda 3.07 2.66 1.15 2.66 7.19 spherical 0 0 0 1 0 1.13

panda 5.33 5.12 1.04 5.12 24 spherical 0 0 0 1 0 2.16

coula 7.6 7.6 1 7.6 36 spherical 0 0 0 1 2 4.5

coula 7.7 7.7 1 7.7 48 spherical 0 2.25 1 1 2 3.38

coula 2.5 2.5 1 2.5 8.27 spherical 0 0 0 1 0 1.54

coula 6.7 6.7 1 6.7 36 spherical 0 0 0 1 2 2.97

coula 7.36 7.36 1 7.36 48 spherical 0 0 0 1 2 3.45

coula 2.88 2.88 1 2.88 7.2 spherical 0 0 0 1 0 1.45

coula 6.15 6.15 1 6.15 27.39 spherical 0 0 0 1 2 2.79

coula 4 4 1 4 13 spherical 0 0 0 1 2 1.74

coula 4.41 4.41 1 4.41 14.55 spherical 1 0 0 1 2 2.15

coula 5.54 5.54 1 5.54 23.23 spherical 0 0 0 1 2 3.79

coula 8.53 8.53 1 8.53 57.11 spherical 0 0 0 1 2 4

coula 10.98 10.98 1 10.98 87.4 spherical 0 0 0 1 2 5.86

coula 8.76 8.76 1 8.76 61.12 spherical 0 0 0 1 2 4.31

coula 7.55 7.55 1 7.55 40.68 spherical 0 0 0 1 2 2.98

coula 5.12 5.12 1 5.12 40 spherical 0 0 0 1 2 3.29

coula 11.34 11.34 1 11.34 90.41 spherical 0 0 0 1 2 5.43

coula 4.73 4.73 1 4.73 18.5 spherical 0 0 0 1 2 2.35

coula 6.35 6.35 1 6.35 33.19 spherical 0 0 0 1 2 3.28

coula 5.09 5.09 1 5.09 30.3 spherical 0 0 0 1 2 2.56

coula 5.81 5.81 1 5.81 24 spherical 1 0 0 1 2 3.15

coula 6.76 6.76 1 6.76 32.16 spherical 1 0 0 1 2 3.28

coula 5.43 5.43 1 5.43 22.79 spherical 0 0 0 1 2 2.56

coula 4.71 4.71 1 4.71 19.28 spherical 0 0 0 1 2 2.07

coula 5.18 5.18 1 5.18 22.92 spherical 0 0 0 1 2 2.47

coula 7.58 7.58 1 7.58 42.52 spherical 0 0 0 1 2 4.13

coula 3.29 3.29 1 3.29 14.59 spherical 0 0 0 1 2 1.69

coula 6.35 6.35 1 6.35 37.24 spherical 0 0 0 1 2 2.97

coula 5.36 5.36 1 5.36 25.06 spherical 0 0 0 1 2 2.66

coula 5.23 5.23 1 5.23 18.26 spherical 1 0 0 1 2 2

coula 5.45 5.45 1 5.45 25.54 spherical 1 0 0 1 2 2.67

coula 4.13 4.13 1 4.13 15.29 spherical 0 0 0 1 2 1.54

193
coula 7.88 7.88 1 7.88 46.8 spherical 0 0 0 1 2 3.38

coula 9.34 9.34 1 9.34 66.9 spherical 0 0 0 1 2 4.9

coula 6.64 6.64 1 6.64 30.14 spherical 1 0 0 1 2 2.56

coula 7.81 7.81 1 7.81 44.41 spherical 0 1.14 1 1 2 3.6

coula 3.44 3.44 1 3.44 9.67 spherical 0 0 0 1 0 1.44

coula 6.47 6.47 1 6.47 30.62 spherical 0 0 0 1 2 3.19

coula 6.88 6.88 1 6.88 32.11 spherical 0 1.14 1 1 2 2.76

coula 6.66 6.66 1 6.66 34.56 spherical 0 1.14 1 1 2 3.69

coula 4.15 4.15 1 4.15 13.52 spherical 0 1.62 2 1 2 1.85

coula 7.91 7.91 1 7.91 46.31 spherical 0 0 0 1 2 4.51

coula 5.92 5.92 1 5.92 24.25 spherical 1 0 0 1 2 2.46

coula 3.75 3.75 1 3.75 16.7 spherical 1 0 0 1 2 1.84

coula 9.5 8.18 1.16 8.18 64.16 spherical 1 0 0 1 2 3.69

coula 6.45 6.14 1.05 6.14 30.29 spherical 1 0 0 1 1 2.15

coula 8.88 8.51 1.04 8.51 55.31 spherical 0 0 0 1 2 3.07

coula 5.73 5.16 1.11 5.16 21.69 spherical 2 0 0 1 1 1.45

coula 3.9 3.43 1.14 3.43 9.36 oblate 2 0 0 1 0 1.65

conovoid
coula 8.66 8.41 1.03 8.41 56.62 spherical 0 0 0 1 2 2.97

coula 6.63 6.51 1.02 6.51 32.19 spherical 0 0 0 1 2 2.46

napoleona 6.32 4.17 1.52 4.17 23.45 ovoid 0 0 0 1 1 2.26

napoleona 5.44 3.66 1.49 3.66 12 elongate ovoid 0 0 0 1 0 1.97

napoleona 5.84 5.64 1.04 5.64 20.17 spherical 0 0 0 1 0 2.06

napoleona 5.08 4.65 1.09 4.65 16.55 ovoid 0 0 0 1 0 1.85

napoleona 4.92 4.53 1.09 4.53 17.7 spherical 0 0 0 1 0 2.25

napoleona 4.76 4.63 1.03 4.63 15.39 spherical 2 0 0 1 0 1.81

napoleona 7.12 5.65 1.26 5.65 32.47 spherical 0 0 0 1 0 3.08

napoleona 6.36 4.71 1.35 4.71 22.32 spherical 0 0 0 1 0 2.11

napoleona 4.72 3.28 1.44 3.28 15.57 spherical 0 0 0 1 0 1.99

napoleona 3.55 2.69 1.32 2.69 7.14 spherical 1 0 0 1 0 1.54

napoleona 5.15 5.08 1.01 5.08 22.96 prolate 0 0 0 1 0 2.56

napoleona 4.75 3.41 1.39 3.41 11.63 ovoid 1 0 0 1 0 1.88

napoleona 4.22 4.17 1.01 4.17 15.78 ovoid 1 0 0 1 0 1.55

napoleona 3.53 3.16 1.12 3.16 7.8 polygon 0 0 0 1 0 2.44

napoleona 5.51 5.34 1.03 5.34 25.52 spherical 0 0 0 1 0 2.61

napoleona 4.35 3.44 1.26 3.44 13.89 ovoid 0 0 0 1 0 1.54

napoleona 4.96 3.73 1.33 3.73 18.9 triangular 1 0 0 2 0 2.25

napoleona 5.31 4.1 1.3 4.1 16.47 triangular 1 0 0 2 0 2.21

napoleona 3.79 3.5 1.08 3.5 12 prolate 0 0 0 1 0 2.39

napoleona 6.66 5.32 1.25 5.32 25.8 prolate 0 0 0 1 0 2.66

napoleona 5.71 5.26 1.09 5.26 23.09 ovoid 1 0 0 1 0 2.25

napoleona 5.31 4.86 1.09 4.86 20.11 spherical 0 0 0 1 0 1.98

napoleona 4.78 4.71 1.01 4.71 19.64 spherical 1 0 0 1 0 2.09

napoleona 6.27 4.61 1.36 4.61 22.13 elongate ovoid 0 0 0 1 0 2.41

napoleona 6.49 6.07 1.07 6.07 31.17 spherical 0 0 0 1 0 2.76

napoleona 4.61 4.53 1.02 4.53 14.91 spherical 1 0 0 1 0 2.19

napoleona 5.07 4.63 1.1 4.63 20.34 spherical 0 0 0 1 0 1.96

194
napoleona 4.51 3.69 1.22 3.69 13.93 spherical 0 0 0 1 0 1.64

napoleona 4.49 3.77 1.19 3.77 12.72 spherical 1 0 0 1 0 1.69

napoleona 3.99 3.89 1.03 3.89 12.09 spherical 0 0 0 1 0 1.54

napoleona 6.14 4 1.54 4 19.5 prolate 0 0 0 1 1 2.77

napoleona 5.02 4.61 1.09 4.61 19.58 prolate 0 0 0 1 0 2.66

napoleona 5.53 5.22 1.06 5.22 22.7 spherical 1 0 0 1 0 1.74

napoleona 7.2 2.9 2.48 2.9 15.98 triangular 1 0 0 2 0 2.36

napoleona 6.45 3.71 1.74 3.71 18.98 quadrangular 2 0 0 2 0 3.11

napoleona 6.96 4.29 1.62 4.29 22.36 quadrangular 2 0 0 2 0 3.52

napoleona 5.04 4.05 1.24 4.05 16.02 polygon 2 0 0 1 0 1.85

napoleona 7.71 5.43 1.42 5.43 33.27 elongate ovoid 0 0 0 1 0 4.03

napoleona 7.56 5.69 1.33 5.69 31.76 ovoid 1 0 0 1 0 4.2

napoleona 3.88 3.27 1.19 3.27 9.22 prolate 0 0 0 1 0 1.47

napoleona 2.67 2.46 1.09 2.46 6.5 spherical 0 0 0 1 0 1.13

napoleona 2.66 2.56 1.04 2.56 5.5 spherical 0 0 0 1 0 1.13

napoleona 2.66 2.56 1.04 2.56 2.89 spherical 1 0 0 1 0 0.72

napoleona 8.79 6.74 1.3 6.74 47.3 ovoid 2 0 0 1 0 3.28

napoleona 8.35 4.75 1.76 4.75 31.01 elongate ovoid 0 0 0 1 0 3.63

napoleona 7.25 6.25 1.76 6.25 32.95 ovoid 0 0 0 1 0 3.13

napoleona 5.37 4.75 1.76 4.75 18.86 ovoid 0 0 0 1 0 18.86

napoleona 9.06 7.09 1.76 7.09 40.33 ovoid 0 0 0 1 0 4.85

napoleona 11.06 5.14 1.76 5.14 38.99 elongate ovoid 0 0 0 1 0 4.82

napoleona 11.03 4.5 1.76 4.5 38.99 elongate ovoid 0 0 0 1 0 5.79

gilbert 10.26 7.78 1.32 7.78 70.17 ovoid 1 0 0 1 0 3.99

gilbert 9 8.12 1.11 8.12 59.98 ovoid 1 0 0 1 0 2.47

gilbert 13.48 13.05 1.03 13.1 142.87 ovoid 1 0 0 1 0 4.51

gilbert 7.75 7.74 1 5.04 32.76 hemispherical 1 0 0 2 0 2.93

gilbert 14.555 14.33 1.02 1.33 163.84 spherical 1 0 0 1 0 5.44

gilbert 7.27 7.27 1 5.94 39.5 oblate 0 0 0 2 0 1.99

conovoid
gilbert 8.4 8.08 1.04 8.08 62.07 hemispherical 1 0 0 2 0 3.79

gilbert 8.01 6.69 1.2 6.69 49.65 hemispherical 1 0 0 2 0 2.83

gilbert 12.79 10.98 1.16 10.98 119.09 ovoid 1 0 0 1 0 4.51

gilbert 7.18 7.18 1 5.02 29.29 hemispherical 1 0 0 2 0 2.63

gilbert 5.94 5.94 1 5.21 30.51 hemispherical 1 0 0 2 0 2.98

gilbert 4.56 4.56 1 2.42 8.18 hemispherical 1 0 0 2 0 1.19

gilbert 8.7 8.39 1.04 8.39 34.95 hemispherical 1 0 0 2 0 3.73

gilbert 7.34 7.34 1 5.72 65.33 hemispherical 1 0 0 2 0 2.81

gilbert 14.25 11.89 1.2 11.89 133.04 ovoid 2 0 0 1 0 6.04

gilbert 6.78 6 1.13 5.51 30.9 oblate 2 0 0 2 0 3.11

conovoid
gilbert 13.72 11.68 1.17 11.68 131.85 ovoid 2 0 0 1 0 5.14

gilbert 16.01 12.44 1.29 12.4 167.1 ovoid 0 0 0 1 0 7.29

gilbert 10.25 9.22 1.11 9.22 84.46 ovoid 1 0 0 1 0 3.59

gilbert 17 14.56 1.17 14.56 166.92 ovoid 0 0 0 1 0 8.34

gilbert 15.12 13.58 1.11 13.58 154.6 ovoid 1 0 0 1 0 7.76

gilbert 7.74 7.74 1 5.42 30.66 hemispherical 1 0 0 2 0 2.63

195
gilbert 12.54 10.08 1.24 10.08 96.89 prolate 0 4.32 2 1 0 4.22

gilbert 8.31 8.16 1.02 8.16 51.53 oblate 2 0 0 2 0 4.1

conovoid
gilbert 7.16 5.08 1.41 5.08 20.2 oblate 2 0 0 2 0 1.88
conovoid
gilbert 7.62 7.62 1 6.88 34.42 oblate 2 0 0 2 0 3.79
conovoid
gilbert 10.02 9.38 1.07 9.38 75.18 ovoid 1 0 0 1 0 5.03

gilbert 9.44 9.38 1.01 9.38 83.58 ovoid 1 0 0 1 0 5.39

gilbert 12.13 10.86 1.12 10.86 95.93 ovoid 1 0 0 1 0 5.53

gilbert 4.75 4.53 1.05 3 24.19 oblate 2 0 0 2 0 2.34

conovoid
gilbert 9.76 8.54 1.14 8.54 67.71 ovoid 1 0 0 1 0 3.17

gilbert 5.44 5.44 1 2.06 12.16 hemispherical 1 0 0 2 0 1.65

gilbert 10.04 8.57 1.17 8.57 68.53 ovoid 1 0 0 1 0 4.1

gilbert 17.26 14.72 1.17 14.72 205.35 ovoid 2 2.63 1 1 0 7.87

gilbert 18.2 14 1.3 14 196.58 ovoid 1 5 3 1 0 9.74

gilbert 13.08 11.6 1.13 11.6 124.23 ovoid 1 0 0 1 0 6.06

gilbert 8.15 4.45 1.83 4.45 27 hemispherical 1 0 0 2 0 2.7

gilbert 5.49 5.46 1.01 5.46 25.01 spherical 0 0 0 1 0 2.05

gilbert 11.23 9.05 1.24 9.05 89 spherical 0 0 0 1 0 4.73

gilbert 7.84 7.76 1.01 7.76 46.55 oblate 1 0 0 2 0 2.61

conovoid
gilbert 7.18 6.74 1.07 6.74 35.18 oblate 2 0 0 2 0 2.71
conovoid
gilbert 6.64 5.95 1.12 5.95 33.45 spherical 2 0 0 1 0 2.49

gilbert 10.67 10.34 1.03 10.34 99.27 spherical 2 0 0 1 0 5.16

gilbert 7.07 6.37 1.11 6.37 32.19 spherical 2 0 0 1 0 1.97

gilbert 6.86 5.63 1.22 5.63 26.39 spherical 2 0 0 1 0 1.95

gilbert 7.66 6.78 1.13 6.78 41.03 spherical 1 0 0 1 0 1.95

gilbert 6.8 6.74 1.01 6.74 38.36 spherical 1 0 0 1 0 1.57

gilbert 9.13 8.34 1.09 8.34 68.58 hemispherical 1 0 0 2 0 3.07

gilbert 6.83 6.1 1.12 6.1 34.08 hemispherical 1 0 0 2 0 2.1

gilbert 5.29 5.09 1.04 5.09 22.94 hemispherical 1 0 0 2 0 1.54

eremo 3.99 3.99 1 3.99 12.61 spherical 0 0 0 1 0 1.64

eremo 2 2 1 2 3.39 spherical 0 0 0 1 0 0.7

eremo 1.88 1.88 1 1.88 4.2 spherical 0 0 0 1 0 0.74

eremo 4.73 4.73 1 4.73 18.9 spherical 0 0 0 1 0 2.05

eremo 4.65 3.91 1.19 3.91 14.46 prolate 0 0 0 1 0 1.35

eremo 3.89 3.89 1 3.89 11.09 spherical 0 0 0 1 0 1.17

eremo 4.63 4.63 1 4.63 16.53 spherical 0 0 0 1 1 2.05

eremo 3.1 3.1 1 3.1 9 spherical 0 0 0 1 0 1.25

eremo 4.9 4.9 1 4.9 20.14 spherical 0 0 0 1 0 2.65

eremo 2.76 2.76 1 2.76 6.57 spherical 0 0 0 1 0 1.74

eremo 3.38 3.38 1 3.38 9.97 spherical 0 0 0 1 0 1.44

eremo 2.63 2.63 1 2.63 5.17 spherical 0 0 0 1 0 1.13

eremo 4.13 4.13 1 4.13 13.66 spherical 0 0 0 1 0 2.67

eremo 2.58 2.58 1 2.58 5.1 spherical 0 0 0 1 0 1.3

eremo 4.5 4.1 1.1 4.1 14.31 prolate 0 0 0 1 0 1.74

eremo 4.05 3.96 1.02 3.96 12.67 prolate 1 0 0 1 1 2.07

196
eremo 3.05 3.05 1 3.05 9 spherical 0 0 0 1 0 1.65

eremo 4.06 3.83 1.06 3.83 13.63 prolate 0 0 0 1 0 1.85

eremo 3.42 3.42 1 3.42 11 spherical 0 0 0 1 0 1.14

eremo 4.32 4.32 1 4.32 15 spherical 0 0 0 1 0 1.81

eremo 4.12 4.12 1 4.12 14.91 spherical 0 0 0 1 1 1.7

eremo 3.15 3.15 1 3.15 9.46 spherical 0 0 0 1 0 1.43

eremo 3.43 3.43 1 3.43 10.89 spherical 1 0 0 1 0 1.95

eremo 3.89 3.89 1 3.89 12.97 spherical 0 0 0 1 1 1.79

eremo 3.36 3.36 1 3.36 9.65 spherical 1 0 0 1 0 1.45

eremo 2.77 2.77 1 2.77 7.88 spherical 1 0 0 1 0 1.44

eremo 4.43 4.43 1 4.43 15 spherical 0 0 0 1 0 2.06

eremo 2.78 2.78 1 2.78 7.15 spherical 0 0 0 1 0 1.55

eremo 2.58 2.58 1 2.58 5.78 spherical 0 0 0 1 0 0.93

eremo 4.11 4.11 1 4.11 12.89 spherical 1 0 0 1 1 2.05

eremo 2.4 2.4 1 2.4 4.68 spherical 0 0 0 1 0 0.83

eremo 2.89 2.89 1 2.89 7.61 spherical 1 0 0 1 0 1.14

eremo 3.07 3.07 1 3.07 7.23 spherical 0 0 0 1 0 1.44

eremo 2.39 2.39 1 2.39 4.55 spherical 0 0 0 1 0 1.13

eremo 2.47 2.47 1 2.47 5.08 spherical 0 0 0 1 0 1.02

eremo 4.58 3.86 1.19 3.86 14 prolate 0 0 0 1 0 2.46

eremo 5.14 5.14 1 5.14 21 spherical 0 0 0 1 0 2.25

eremo 2.7 2.7 1 2.7 7 spherical 0 0 0 1 0 1.54

eremo 3.89 3.89 1 3.89 12 spherical 0 0 0 1 0 1.64

eremo 3.69 3.69 1 3.69 10.55 spherical 0 0 0 1 0 1.74

eremo 3.49 3.49 1 3.49 7.6 spherical 0 0 0 1 0 1.55

eremo 3.07 2.17 1.41 2.17 5.64 hemispherical 1 0 0 2 0 1.62

eremo 2.17 2.17 1 2.17 3.43 spherical 0 0 0 1 0 1.03

eremo 7.07 6.37 1.11 6.37 17.69 spherical 0 0 0 1 0 1.95

eremo 6.86 5.63 1.22 5.63 8.51 spherical 0 0 0 1 0 1.23

eremo 7.66 6.78 1.13 6.78 6.27 spherical 0 0 0 1 0 0.92

eremo 6.8 6.74 1.01 6.74 12.3 spherical 0 0 0 1 0 2.15

eremo 9.13 8.34 1.09 8.34 12.35 spherical 0 0 0 1 0 1.84

eremo 6.83 6.1 1.12 6.1 6 spherical 0 0 0 1 0 0.92

eremo 5.29 5.09 1.04 5.09 6.28 spherical 0 0 0 1 0 1.13

calpo 2.6 2.6 1 2.6 5.33 spherical 0 0 0 1 0 1.25

calpo 2.16 1.75 1.23 1.75 3.49 ovoid 0 0 0 1 0 0.72

calpo 2.36 2.36 1 2.36 4.13 spherical 0 0 0 1 0 0.83

calpo 2.66 2.66 1 2.66 6 spherical 0 0 0 1 0 1.52

calpo 1.84 1.84 1 1.84 3.13 spherical 0 0 0 1 0 1.33

calpo 2.29 2.29 1 2.29 4.4 spherical 0 0 0 1 0 1.33

calpo 2.27 2.27 1 2.27 4 spherical 0 0 0 1 0 0.7

calpo 1.81 1.81 1 1.81 3.34 spherical 0 0 0 1 0 0.93

calpo 1.69 1.69 1 1.69 2.3 spherical 0 0 0 1 0 0.94

calpo 2.15 2.15 1 2.15 3.12 spherical 0 0 0 1 0 0.93

calpo 2.66 2.66 1 2.66 3.52 spherical 0 0 0 1 0 1.59

calpo 2.43 2.43 1 2.43 3.51 spherical 0 0 0 1 0 0.93

197
calpo 1.92 1.92 1 1.92 3.29 spherical 0 0 0 1 0 1.23

calpo 2.16 2.16 1 2.16 4.6 spherical 0 0 0 1 0 1.13

calpo 2.05 2.05 1 2.05 3.31 spherical 1 0 0 1 0 1.05

calpo 2.87 2.87 1 2.87 6.4 spherical 0 0 0 1 0 1.33

calpo 2.05 2.05 1 2.05 4 spherical 0 0 0 1 0 0.92

calpo 1.96 1.96 1 1.96 3.43 spherical 0 0 0 1 0 0.61

calpo 2.35 2.35 1 2.35 4.35 spherical 0 0 0 1 0 0.92

calpo 2.15 2.15 1 2.15 5.1 spherical 0 0 0 1 0 0.83

calpo 1.95 1.95 1 1.95 4.08 spherical 0 0 0 1 0 0.72

calpo 2.25 2.25 1 2.25 6.11 spherical 0 0 0 1 0 1.11

calpo 2.25 2.25 1 2.25 4.91 spherical 0 0 0 1 0 1.03

calpo 2.46 2.46 1 2.46 4.59 spherical 0 0 0 1 0 1.03

calpo 1.95 1.95 1 1.95 3.78 spherical 0 0 0 1 0 0.93

calpo 2.35 2.35 1 2.35 4.23 spherical 0 0 0 1 0 0.92

calpo 2.85 2.85 1 2.85 8.51 spherical 0 0 0 1 0 1.85

calpo 2.36 2.36 1 2.36 4.51 spherical 0 0 0 1 0 1.23

calpo 1.95 1.95 1 1.95 4.3 spherical 1 0 0 1 0 0.72

calpo 1.65 1.65 1 1.65 2.28 spherical 0 0 0 1 0 0.72

calpo 2.57 2.57 1 2.57 5.33 spherical 0 0 0 1 0 1.2

calpo 2.76 2.76 1 2.76 6.52 spherical 0 0 0 1 0 1.33

calpo 2.77 2.77 1 2.77 6.09 spherical 0 0 0 1 0 1.57

calpo 1.45 1.45 1 1.45 1.98 spherical 0 0 0 1 0 0.72

calpo 2.46 2.46 1 2.46 5.3 spherical 0 0 0 1 0 0.92

calpo 1.95 1.95 1 1.95 2.6 spherical 0 0 0 1 0 0.82

calpo 2.15 2.15 1 2.15 3.52 spherical 0 0 0 1 0 0.83

calpo 1.74 1.74 1 1.74 2.46 spherical 0 0 0 1 0 0.61

calpo 2.98 2.98 1 2.98 5.44 spherical 0 0 0 1 0 1.02

calpo 1.95 1.95 1 1.95 4.11 spherical 0 0 0 1 0 1.2

calpo 2.07 2.07 1 2.07 3.96 spherical 0 0 0 1 0 1.04

calpo 2.46 2.05 1.2 2.05 5.87 hemispherical 1 0 0 2 0 1.02

calpo 2.53 2.05 1.23 2.05 6.1 ovoid 0 0 0 1 0 1

calpo 2.96 2.38 1.24 2.38 5.72 prolate 0 0 0 1 0 1.39

calpo 2.82 2.82 1 2.82 6.15 spherical 0 0 0 1 0 1.41

calpo 2.31 2.24 1.03 2.24 5.34 hemispherical 1 0 0 1 0 1.15

calpo 2.76 1.94 1.42 1.94 4.03 hemispherical 1 0 0 1 0 1.38

calpo 2.68 2.46 1.09 2.46 3.86 hemispherical 1 0 0 1 0 1.34

calpo 2.96 2.1 1.41 2.1 4.45 ovoid 0 0 0 1 0 1.48

calpo 2.5 2.5 1 2.5 4.84 spherical 0 0 0 1 0 1.25

sarcoph 14.16 14.16 1 12.57 143.83 quadrangular 6 0 0 3 0 5.92

sarcoph 20.9 13.71 1.52 13.71 192.11 polygon 7 0 0 3 0 5.64

sarcoph 17.49 12.45 1.4 12.45 153.67 polygon 8 0 0 3 0 4.4

sarcoph 21.84 15.25 1.43 15.25 294.44 polygon 9 0 0 3 0 6.45

sarcoph 12.56 11.14 1.13 11.14 136.42 polygon 6 0 0 3 0 6.63

sarcoph 14.14 11.62 1.22 11.62 141.63 polygon 4 0 0 3 0 5.94

sarcoph 11.87 9.98 1.19 9.98 113.99 quadrangular 6 0 0 3 0 5.5

sarcoph 11.27 9.98 1.13 9.88 125.26 polygon 7 0 0 3 0 4.66

198
sarcoph 12.47 12.21 1.02 12.21 137.97 polygon 6 0 0 3 0 5.76

sarcoph 11.2 11.1 1.01 11.1 104 polygon 7 0 0 3 0 4.73

sarcoph 7.96 5.83 1.37 5 42.99 quadrangular 6 0 0 3 0 3

sarcoph 5.48 5.04 1.09 4.55 25 quadrangular 5 0 0 3 0 2.3

sarcoph 18.22 12.52 1.46 12.5 176 quadrangular 6 0 0 3 0 5.08

sarcoph 14.68 14.68 1 14.68 147 polygon 7 0 0 3 0 4.03

sarcoph 10.92 9.43 1.16 9.43 72 polygon 7 0 0 3 0 4.45

sarcoph 10.08 8.76 1.15 8.76 82 quadrangular 6 0 0 3 0 2.91

sarcoph 11.45 10.45 1.1 10.45 116 quadrangular 6 0 0 3 0 6.38

sarcoph 13.56 11.33 1.2 11.33 135.67 polygon 7 0 0 3 0 3.69

sarcoph 16.49 9.73 1.69 9.73 148 polygon 6 0 0 3 0 5.45

concaveconve
x
sarcoph 12.58 12.23 1.03 12.23 120.81 polygon 7 0 0 3 0 6

sarcoph 12.23 10 1.22 10 96.81 polygon 7 0 0 3 0 7.44

sarcoph 14.37 11.21 1.28 11.21 119.21 polygon 6 0 0 3 0 10.01

sarcoph 12.36 11.77 1.05 11.77 117.06 hemispherical 1 0 0 2 0 6.24

sarcoph 12.5 10.85 1.15 10.85 10.85 hemispherical 3 5.03 4 2 0 5.16

sarcoph 17 10.04 1.69 10.4 131.55 polygon 9 0 0 3 0 6.1

sarcoph 14.87 10.67 1.39 10.67 141.83 polygon 7 0 0 3 0 5.52

sarcoph 9.88 8.25 1.2 8.25 85.73 polygon 7 0 0 3 0 4.92

sarcoph 10.04 6.69 1.5 6.69 70.08 polygon 7 0 0 3 0 4.85

sarcoph 10.82 9.14 1.18 9.14 85.49 polygon 7 0 0 3 0 4.2

sarcoph 9.91 8.44 1.17 8.44 88.71 polygon 7 0 0 3 0 4.9

sarcoph 10.86 9.58 1.13 9.58 91.81 polygon 7 0 0 3 0 4.72

sarcoph 10.2 9.13 1.12 9.13 91.65 polygon 6 0 0 3 0 5.27

sarcoph 15.69 9.65 1.63 9.65 130.79 polygon 7 0 0 3 0 6.16

concaveconve
x
sarcoph 10.7 8.03 1.33 8.03 79.75 polygon 6 0 0 3 0 4.64

sarcoph 10.86 9.73 1.12 9.76 85.53 polygon 7 0 0 3 0 5.754

sarcoph 8.78 5.57 1.58 5.57 51.99 polygon 6 0 0 3 0 4.45

sarcoph 20.15 15.79 1.28 5.57 308.59 polygon 6 0 0 3 0 9.24

sarcoph 20.81 16.35 1.27 5.57 232.32 polygon 7 0 0 3 0 6.96

sarcoph 23.75 12.01 1.98 5.57 181.22 polygon 7 0 0 3 0 6.07

sarcoph 17.1 14.59 1.17 5.57 205.63 polygon 9 0 0 3 0 13.16

sarcoph 19.99 13.95 1.43 5.57 163.25 polygon 9 0 0 3 0 5.95

sarcoph 16.41 12.4 1.32 5.57 193.32 polygon 6 0 0 3 0 9.76

sarcoph 17.72 15 1.18 5.57 275.54 polygon 7 0 0 3 0 9.31

sarcoph 18.26 16.95 1.08 7.53 61.53 polygon 6 0 0 3 0 3.58

sarcoph 11 9.09 1.21 9.09 69.42 polygon 6 0 0 3 0 3.62

sarcoph 9.63 9.45 1.02 9.45 82.19 polygon 7 0 0 3 0 4.92

sarcoph 12.47 10.5 1.19 10.5 102.83 polygon 7 0 0 3 0 4.7

sarcoph 9.01 8.87 1.02 8.87 55.17 polygon 7 0 0 3 0 4.83

sarcoph 9.41 7.81 1.2 7.81 60.64 polygon 5 0 0 3 0 4.5

sarcoph 20.15 15.79 1.28 10 258.38 polygon 9 0 0 3 0 9.77

xylia 6.27 5.27 1.19 5.27 26.26 prolate 0 0 0 1 0 1.77

xylia 2.98 2.7 1.1 2.7 6.67 prolate 0 0 0 1 0 0.93

199
xylia 4.27 3.69 1.16 3.69 13.63 prolate 0 0 0 1 0 1.23

xylia 2.9 2.26 1.28 2.69 5.15 hemispherical 1 0 0 2 0 1.13

xylia 2.81 2.59 1.08 2.59 5.35 hemispherical 1 0 0 2 0 1.3

xylia 4.05 2.7 1.5 2.7 8.46 ovoid 0 0 0 1 0 1.47

xylia 3.38 3.07 1.1 3.07 8.27 prolate 1 0 0 1 0 2.27

xylia 2.87 2.87 1 2.87 7 spherical 0 0 0 1 0 1.03

xylia 4.15 3.69 1.12 3.69 11.74 ovoid 0 0 0 1 0 1.79

xylia 2.77 2.66 1.04 2.66 5 prolate 0 0 0 1 0 0.93

xylia 5.34 3.62 1.48 3.62 14.56 ovoid 0 0 0 1 0 1.96

xylia 3.38 3.28 1.03 3.28 10.53 spherical 0 0 0 1 0 1.62

xylia 5.2 5.2 1 5.2 19 spherical 0 0 0 1 0 2.32

xylia 5.25 4.49 1.17 4.49 16.75 ovoid 0 0 0 1 0 1.64

xylia 4.78 4.78 1 4.78 20 spherical 1 0 0 1 0 1.85

xylia 3.38 3.07 1.1 3.07 8.9 prolate 0 0 0 1 0 1.64

xylia 4.76 3.71 1.28 3.71 14.12 prolate 1 0 0 1 0 1.89

xylia 4.22 4.13 1.02 4.13 15.62 ovoid 2 0 0 1 0 2.17

xylia 3.4 3.18 1.07 3.18 9.08 spherical 1 0 0 1 0 1.65

xylia 5.6 4.65 1.2 4.65 21.12 prolate 1 0 0 1 0 2.29

xylia 4.45 3.44 1.29 3.44 11.84 ovoid 1 0 0 1 0 1.42

xylia 4.18 4.18 1 4.18 13.66 spherical 0 0 0 1 0 1.44

xylia 2.99 2.64 1.13 2.64 6.71 prolate 1 0 0 1 0 1.02

xylia 3.2 3.2 1 3.2 8.57 spherical 1 0 0 1 0 1.85

xylia 3.83 2.11 1.82 2.11 10.05 hemispherical 1 0 0 2 0 1.45

xylia 4.06 3.21 1.26 3.21 10.82 hemispherical 1 0 0 2 0 2.25

xylia 5.31 4.12 1.29 4.12 15.35 prolate 1 0 0 1 0 1.59

xylia 5.77 5.04 1.14 5.04 23.53 prolate 1 0 0 1 0 2.16

xylia 4.22 3.21 1.31 3.21 10.42 ovoid 0 0 0 1 0 1.64

xylia 6 4.51 1.33 4.51 20.11 ovoid 1 0 0 1 0 2.11

xylia 5.13 2.98 1.72 2.98 15.29 hemispherical 1 0 0 2 0 1.74

xylia 5.45 4.18 1.3 4.18 17.82 prolate 1 0 0 1 0 3.18

xylia 2.62 2.62 1 2.62 5.76 spherical 0 0 0 1 0 0.82

xylia 2.2 2.2 1 2.2 4 spherical 0 0 0 1 0 1

xylia 4.93 4.43 1.11 4.43 17 spherical 1 0 0 1 0 2.11

xylia 6.21 5.51 1.13 5.51 29.98 prolate 1 0 0 1 0 2.05

xylia 4.64 3.55 1.31 3.55 11.17 ovoid 1 0 0 1 0 0.72

xylia 5.45 4.03 1.35 4.03 16.51 ovoid 1 0 0 1 0 1.13

xylia 3.84 3.26 1.18 3.26 11.71 prolate 1 0 0 1 0 1.55

xylia 6.56 3.43 1.91 3.43 6.56 prolate 2 0 0 2 0 1.3

xylia 4.6 4.34 1.06 4.34 17.95 prolate 1 0 0 1 0 1.95

xylia 4.61 3.78 1.22 3.78 14.33 prolate 0 0 0 1 0 1.74

xylia 3.92 3.13 1.25 3.13 10.24 ovoid 2 0 0 1 0 1.39

xylia 5.62 5.43 1.03 5.43 25.45 ovoid 1 0 0 1 0 1.84

xylia 4.5 3.33 1.35 3.33 10 ovoid 0 0 0 1 0 0.94

xylia 5.02 4.7 1.07 4.7 22.2 spherical 2 0 0 1 0 1.55

xylia 2.78 2.36 1.18 2.36 8.15 spherical 2 0 0 1 0 1.33

xylia 2.97 2.46 1.21 2.46 6 spherical 1 0 0 1 0 1.33

200
xylia 3.09 2.87 1.08 2.87 7.75 spherical 1 0 0 1 0 1.57

xylia 3.85 3.25 1.18 3.25 11.43 spherical 1 0 0 1 0 1.25

treculia 8.91 5.07 1.76 5.07 39.15 ovoid 0 0 0 1 0 3.63

treculia 7.5 4.92 1.52 4.92 25.04 oblate 0 0 0 1 0 3.4

conovoid
treculia 8.38 6.99 1.2 6.99 38.74 ovoid 1 0 0 1 0 3.42

treculia 4.81 4.31 1.12 4.31 12.82 triangular 3 0 0 2 0 1.9

treculia 6.97 5.21 1.34 5.21 29.15 ovoid 0 0 0 1 0 2.53

treculia 11.6 8.22 1.41 8.22 66.6 ovoid 0 0 0 1 0 4.1

treculia 7.83 5.74 1.36 5.74 30.67 oblate 0 0 0 1 0 3.04

conovoid
treculia 4.94 4.18 1.18 4.18 14.95 oblate 0 0 0 1 0 2.47
conovoid
treculia 6.91 6.11 1.13 6.11 36.2 ovoid 0 0 0 1 0 2.75

treculia 7.91 5.86 1.35 5.86 29 ovoid 0 0 0 1 0 3.21

treculia 6.52 4.92 1.33 4.92 27.78 ovoid 1 0 0 1 0 2.49

treculia 5.78 4.57 1.26 4.57 23.98 ovoid 1 0 0 1 0 1.65

treculia 8.43 5.91 1.43 5.91 34.9 ovoid 0 0 0 1 0 3.19

treculia 5.93 5.03 1.18 5.03 27.5 ovoid 1 0 0 1 0 2.97

treculia 6.16 4.97 1.24 4.97 22.8 ovoid 0 0 0 1 0 1.95

treculia 5.12 4.81 1.06 4.81 18.45 ovoid 0 0 0 1 0 2.05

treculia 7.41 5.74 1.29 5.74 35.56 ovoid 0 0 0 1 0 2.95

treculia 6.96 5.65 1.23 5.65 31.28 ovoid 0 0 0 1 0 1.87

treculia 15.73 8.65 1.82 8.65 112.17 pyriform 0 0 0 1 0 6.65

treculia 6.69 4.57 1.46 4.57 21.49 oblate 0 0 0 1 0 2.66

conovoid
treculia 7.36 5 1.47 5 29.18 pyriform 0 0 0 1 0 2.9

treculia 4.99 4.36 1.14 4.36 20.12 spherical 1 0 0 1 0 1.88

treculia 6.7 6.42 1.04 6.42 33.52 ovoid 0 0 0 1 0 3.08

treculia 6.92 5.28 1.31 5.28 28.34 ovoid 0 0 0 1 0 2.66

treculia 9.03 7.33 1.23 7.33 50.23 ovoid 0 2.03 2 1 0 2.82

treculia 9.42 7.99 1.18 7.99 59.23 plano-convex 0 1.85 1 1 0 3.66

treculia 11.27 7.33 1.54 7.33 67.5 ovoid 0 0 0 1 0 4.77

treculia 6.7 6.05 1.11 6.05 31.26 ovoid 0 0 0 1 0 3.35

treculia 8.46 6.53 1.3 6.53 39.46 ovoid 0 0 0 1 0 2.25

treculia 7.67 6.64 1.16 6.64 35.62 prolate 0 0 0 1 0 2.77

treculia 6.65 5.32 1.25 5.32 28.98 ovoid 1 0 0 1 0 2.66

treculia 10.58 7.47 1.42 7.47 60.42 ovoid 0 0 0 1 0 4.55

treculia 5.68 5.31 1.07 5.31 23.85 hemispherical 1 0 0 1 0 2.07

treculia 6.43 5.09 1.26 5.09 24.92 ovoid 0 0 0 1 0 1.85

treculia 8.99 6.33 1.42 6.33 46.95 ovoid 0 0 0 1 0 4.54

treculia 7.97 5.58 1.43 5.58 33.01 prolate 0 0 0 1 0 2.44

treculia 6.11 5.74 1.06 5.74 28.51 hemispherical 0 0 0 1 1 2.93

treculia 7.15 5.47 1.31 5.47 31.15 ovoid 1 0 0 1 0 3.21

treculia 7.34 5.6 1.31 5.6 32.68 ovoid 0 0 0 1 0 3.2

treculia 8.17 4.98 1.64 4.98 33.67 plano-convex 0 0 0 1 0 3.98

treculia 3.08 2.97 1.04 2.97 7.5 spherical 0 0 0 1 0 1.23

treculia 3.29 2.97 1.11 2.97 8.41 spherical 0 0 0 1 0 1.43

treculia 7.17 5.19 1.38 5.19 26.68 ovoid 0 7.7 5.84 1 0 2.83

201
treculia 12.47 10.07 1.24 10.07 89.85 ovoid 0 0 0 1 0 5.85

treculia 7.23 7.06 1.02 7.06 37.1 spherical 0 0 0 1 1 3.48

treculia 13.32 10.14 1.31 10.14 93.77 ovoid 0 0 0 1 0 6.33

treculia 5.96 4.1 1.45 4.1 20.23 hemispherical 0 0 0 1 0 2.98

treculia 5.04 4.42 1.14 4.42 18.39 hemispherical 0 0 0 1 0 2.52

treculia 9.73 6.58 1.48 6.58 47.73 ovoid 0 0 0 1 0 3.91

treculia 9.63 8.5 1.13 8.5 58.52 prolate 1 0 0 1 0 4.1

Appendix table 7: Microremain variables used for identification model.

Variable Description Metric

Shared variables
Length Maximum diameter (µm), measured from spine tip to spine tip Numeric
(µm)
Width Maximum diameter (µm) perpendicular to the maximum diameter Numeric
(µm)
LW Ratio Length to width ratio Numeric
(µm)
Area Total observable area in a 2D plane Numeric
(µm2)
Shape Ovoid, elongate ovoid, pyriform, oblate conovoid, elongate conovoid, hemispherical, 16
triangular, quadrangular, polygon, polygon concave-convex, angularpoint, angulate descriptors
elongate, ovoid concave-convex, prolate concave
Starch specific
Facets Total number of maximum observable facets Counts
Lam Lamellae presence and distinctness 0-3 scale
Dist Distance of longest arm of cross observed on cross-polarised light Numeric
Striaelen Average length of radial striae/cracks visible on the starch Numeric
Striaeno Number of radial striae/cracks visible on the starch Counts
Type simple, semi-compound or compound classification 3 descriptors
Phytolith specific
Irregul Measure of phytolith surface irregularity 0-4 scale
Spinelen Estimated mean spine length: the mean length of spines approximately parallel with the Numeric
viewing plane (µm)
Spineno Number of spines visible in entirety in the viewing field. Spines were counted value if Numeric
their base was not obscured by the phytolith.
Conjoined Score of phytolith attachment to other phytoliths 1-2 scale

Appendix table 8: Random forest phytolith identification model. Using spheroid, globular morphotypes only.
Identification rate=rate of successful identification per genus.

Number of variables tried at each split (mtry) 15

Tune length 3
Tree number 500
Out of bag estimate of error rate 25.75 %
Confusion matrix
Aframomum Ancistrophyllum Elaeis Eremospatha Sarcophrynium Identification
rate
Aframomum 39 3 1 5 2 0.78
Ancistrophyllum 3 32 3 12 0 0.64
Elaeis 2 3 40 5 0 0.8

202
Eremospatha 5 11 1 33 0 0.66
Sarcophrynium 2 0 1 0 47 0.94

Appendix table 9: Random forest starch identification model. Identification rate=rate of successful
identification per genus.

Number of variables tried at each split (mtry) 14

Tune length 3
Tree number 500
Out of bag estimate of error rate 32.77 %
Confusion matrix
Aframomum

Calpocalyx

Cola

Coula

Eremospatha

Gilbertiodendron

Napoleona

Panda

Piper

Sacoglottis

Sarcophrynium

Treculia

Xylia

rate
Identification
Aframomum 45 1 0 0 0 0 0 0 2 0 0 1 1 0.9
Calpocalyx 0 40 0 0 7 0 0 2 0 0 0 0 1 0.8
Cola 0 0 26 0 0 3 0 2 0 5 0 11 3 0.52
Coula 0 0 0 44 3 0 0 0 0 2 0 0 1 0.88
Eremospatha 0 10 0 0 31 0 0 7 0 0 0 0 2 0.62
Gilbertiodendron 0 0 4 0 0 38 1 0 0 7 0 0 0 0.76
Napoleona 0 2 0 0 1 1 18 7 0 2 0 8 11 0.36
Panda 0 3 1 0 6 0 11 11 0 0 0 6 12 0.22
Piper 2 0 0 0 0 0 0 0 47 1 0 0 0 0.94
Sacoglottis 0 0 0 0 0 6 0 0 0 43 0 0 1 0.86
Sarcophrynium 0 0 0 0 0 2 0 0 1 0 47 0 0 0.94
Treculia 0 0 7 0 0 2 4 6 0 1 0 26 4 0.52
Xylia 0 3 0 0 6 0 7 7 0 3 0 3 21 0.42

203
Appendix table 10: All recovered microremains in each dental calculus sample. M=many.

Tina

Agathe
Rubra

Mkubwa

Clyde

Kendo

Leo

Lefkas

Zerlina

Castor

Fanny

Goma

Hector

Brutus
Noah

Ondine

Venus

Bijou

Dorry

Oreste

13438
m
Loukou

Piment
o
Leonard

Bambou

Ophelia
Chimp

Starches 4 9 7 - 4 - 5 1 - 2 54 16 2 3 5 - 16 1 4 2 15 1 - - - -
1 5 7 0
Possible starches 4 4 3 - - - - - - - - 14 0 11 1 - 2 - 1 2 - 2 - - - 1
Phytoliths Spheroid echinate 14 71 9 4 18 18 100 1 11 5 98 70 2 12 3 15 58 7 11 3 69 8 - - - -
0 1 0 0 5 1 9 0 2 2 4
Long cell 3 3 1 1 2 12 9 3 17 - 3 14 - 4 2 1 4 1 9 2 3 3 - - - -
0
Cylindroid 1 - 1 - - 2 1 - - 2 - - - 1 - 1 4 2 1 - - 4 - - - -
Grass short cell - 3 2 - 3 2 1 - - 1 - 1 - 7 2 - 3 - 2 - - 1 - - - -
Hair cell - 2 3 3 - 6 1 - 3 3 1 3 1 6 6 4 7 4 6 - 3 3 - - - -
Acicular hair cell 1 - - - - - - - 1 - 1 2 - - 1 - 3 1 - 2 - - - - - -
Bulliform 3 3 3 - - 10 3 1 1 - 5 3 1 4 1 2 4 2 4 1 1 1 - - - -
Parallepipedal - 2 3 1 2 11 5 2 6 - - 1 - 4 2 1 5 3 7 - 2 - - - - -
Plate 1 - 1 - - 2 1 1 1 - - 1 - 2 - - 1 - 2 - - - - - - -
Undenti. phytolith 6 9 7 2 1 7 5 2 7 4 - 2 1 3 3 1 7 2 15 1 2 2 - - - -
Tracheid - 1 - - 1 - - - - - - - - - - - - - 1 - 1 - - - - -
Ellipsoid - - - - - - - - 1 - 1 1 - 1 - 1 1 - - - 1 - - - - -
Unsilicified plants cells Monocot - 1 1 - - - - - 3 1 - 2 1 3 - - - - - - - - - - - -
Dicot - 3 - - - - - - 2 - - - - - - - - 1 - - 2 - - - - -
Unclear 6 10 1 - - 6 1 6 3 3 7 4 2 14 1 1 2 1 3 1 4 1 - - 1 -
3 4 2
Stoma - - - - - 1 - - - - 1 - - - - - - 1 - - - - - - - -
Dicot stoma - - - - - - - - - - - - - - 1 9 - - - - - 1 - - - -
Palm - - - - - - - - - - 2 - - 2 - - - - 1 - - - - - - -
Spiral thickening - - 1 - - - - - - - - 3 - - - - - 1 - - 2 - - - - -
3
Honeycomb sheet - - - - - 2 - - - - - 2 - 3 2 - - 1 - - 1 - - - - -
Stellate hair - - - - - 1 - - - - - - - 2 - - - - - - 1 - - - - -
Hairs - 8 1 1 3 14 8 7 1 2 1 2 2 1 6 - - 2 - - 3 - - - - -
 Jigsaw - - 2 - - - - - - - - 3 - - 9 - - 4 - - - - - - - -
Tracheid - - - - - 1 - - - - - 1 1 - - - - - - - - - - - - -
Fungal spore 1 - M - - - - M M M - M M M M - - - - - M M - - - 1

Diatom - - - 1 - - 1 1 - 5 1 1 1 1 - - 1 2 - - 1 - - - - -
Pollen - - - - - - 1 - - - 5 1 3 - - - 2 5 2 - 2 - - - - -
Cystolith - - - - - - - - - - - - - - - - 1 1 - - 1 - - - - -
Barbule - - - - - - - - - - 2 - - - 1 - - - - - 2 - - - - -
Indeterminate rod 1 - - - - - - - - - - 8 - - - - - - - - - - - - - -
Possible starch amyloplast (Aframomum?) - - - - - - - - - - - 1 - - - - - - - - - - - - - -
Oxalate 15 -
Falcates - - - - - - 6 - - - - - - - - - - - - - - - - - - -
Cup hair - - - - - 9 - - - - - - - - - - - - - - - - - - - -
Feather hair - - - - - 1 - - - - - - - - - - - - - - - - - - - -
Mammal hair - 1 - - - - - - - - - - - - - - - - - - - - - - - -
Insects - - - - - - - - 1 7 2 - - - - - - - 1 - 1 - - - - -
Insect hairs - - - - - - - 1 - - - - - - - - - 1 - - - - - - - -
insect scale - - - - - - - - - 1 - 1 - - - - - 1 - - - - - - - -
Unknown 4 5 7 2 - - 4 3 - 5 3 2 2 5 4 - 1 2 1 1 2 - - - - -

Phytoliths included in identification model 3 6 7 - 3 - 5 8 - 2 38 10 2 4 - - 1 9 4 2 N/ N/ - - - -

2 3 6 A A
Starches included in identification model 9 52 1 3 9 16 94 1 76 6 12 12 2 93 2 9 4 6 86 3 N/ N/ - - - -
0 1 9 4 9 1 7 7 3 5 A A
4

205
Appendix table 11: Counts of identified genera in Taï Chimpanzee calculus samples.

Phytolith Starch
Name Genera % of total genera Genera count % of total genera
count
Ophelia 0 0 0 0
Leonardo 0 0 0 0
Bambou 0 0 0 0
Piment 0 0 0 0
Oreste 5 100 2 15.38
Hector 3 60 2 15.38
Noah 5 100 0 0
Lefkas 2 40 4 30.77
Tina 3 60 2 15.38
Dorry 4 80 3 23.08
Zerlina 4 80 0 0
Clyde 3 60 3 23.08
Agathe 4 80 4 30.77
Bijou 5 100 5 38.46
Leo 4 80 2 15.38
Castor 5 100 3 23.08
Fanny 4 80 10 76.92
Kendo 5 100 0 0
Venus 4 80 5 38.46
Goma 5 100 9 69.23
Rubra 5 100 5 38.46
Ondine 3 60 0 0
Mkubwa 2 40 0 0
Brutus 5 100 3 23.08

Appendix table 12: Measurements of phytoliths from calculus. ER=Eremospatha, AF=Aframomum,

AN=Laccosperma, EL=Elaeis, SA=Sarcophrynium.
name
Chimpanzee

Length

Width

LW Ratio

Brea

Area

Irregul

Spinelen

Spineno

Spineang

Shape

Conjoined

Plant genera
score
Certainty

Leo 7.89 7.68 1.03 7.68 48.0 3 0.92 6 99 spherical 1 ER 0.53

Leo 9.74 9.43 1.03 9.43 79.6 3 0.92 10 91 spherical 1 ER 0.49

Leo 6.59 4.18 1.58 4.18 21.7 3 0.65 8 94 ovoid 1 ER 0.60

Leo 8.5 6.41 1.33 6.41 46.2 4 0.51 9 110 ovoid 1 AF 0.42

Leo 6.67 6.39 1.04 6.39 36.6 3 0.88 8 98 spherical 1 ER 0.77

Leo 3.1 2.71 1.14 2.71 7.4 2 0.4 5 88 polygon 1 ER 0.56

Leo 6.84 5.74 1.19 5.74 33.0 2 0.75 6 110 spherical 1 ER 0.77

Leo 4.72 4.72 1 4.72 19.2 4 0.91 8 82 polygon 1 ER 0.40

Leo 10.49 9.08 1.16 9.08 76.6 3 1.04 6 77 spherical 1 AN 0.56

Leo 12.01 8.71 1.38 8.71 87.6 2 0.8 7 131 prolate 1 EL 0.41

Leo 13.63 13.42 1.02 13.42 140.7 3 1.33 13 95.81 spherical 1 EL 0.77

Leo 5.51 4.58 1.2 4.58 22.8 2 0.4 5 113 spherical 1 ER 0.61

Leo 7.97 3.42 2.33 3.42 28.1 4 0.7 10 80 prolate 1 EL 0.69

Leo 9.56 7.17 1.33 7.17 45.6 4 0.87 8 90 ovoid 1 ER 0.53

Leo 6.4 4.92 1.3 4.92 28.8 2 0.55 7 106 spherical 1 ER 0.53

Leo 5.03 4.22 1.19 4.22 16.8 2 0.66 7 113 spherical 1 ER 0.94

Leo 17.86 11.05 1.62 11.05 147.0 2 0.58 14 103 ovoid 1 EL 0.65

Leo 11.28 10.86 1.04 10.86 98.2 2 0.88 15 88 spherical 1 EL 0.77

Leo 8.64 6.45 1.34 6.45 50.3 3 0.92 7 92 ovoid 1 ER 0.42

Leo 8.55 7.5 1.14 7.5 61.2 4 0.66 7 93 spherical 1 EL 0.39

Leo 5.97 5.07 1.18 5.07 23.8 4 0.9 8 126 polygon 1 ER 0.74

Leo 8.47 7.65 1.11 7.65 52.5 3 0.78 8 89 spherical 1 ER 0.44

Leo 7.93 7.39 1.07 7.39 48.9 2 0.78 11 116 spherical 1 ER 0.39

Leo 10.07 9.63 1.05 9.63 87.8 3 1.28 12 101 spherical 1 EL 0.51

Leo 7.08 5.77 1.23 5.77 34.6 3 0 0 0 polygon 1 AF 0.99

Leo 8.48 6.69 1.27 6.69 44.5 2 0.51 11 94 prolate 1 EL 0.90

Leo 8.76 5.03 1.74 5.03 33.5 3 0.7 10 83 prolate 1 EL 0.56

Leo 8.39 7.79 1.08 7.79 59.0 3 0.87 9 107 spherical 1 ER 0.46

Leo 8.8 8.55 1.03 8.55 62.6 3 0.94 8 98 spherical 1 ER 0.42

Leo 8.66 8.06 1.07 8.06 56.0 3 0.87 8 80 spherical 1 ER 0.38

Leo 11.77 8.89 1.32 8.89 73.0 3 0.97 16 125 prolate 1 EL 0.95

Leo 6.19 5.34 1.16 5.34 28.8 4 0.78 7 72.46 spherical 1 AF 0.52

Leo 9.33 9.02 1.03 9.02 64.5 3 0.87 12 82 spherical 1 EL 0.48

Leo 8.6 8.52 1.01 8.52 58.7 4 0.52 15 81 prolate 1 EL 0.83

Leo 16.1 15.56 1.03 15.56 220.0 2 1 19 87 spherical 1 EL 0.91

Leo 11.98 11.01 1.09 11.01 109.0 2 0.83 15 104 spherical 1 EL 0.94

Leo 10.76 6.92 1.55 6.92 62.0 3 0.83 13 119 ovoid 1 EL 0.97

Leo 10.31 9.78 1.05 9.78 80.2 3 0.75 10 80 spherical 1 ER 0.41

Leo 16.43 15.22 1.08 15.22 283.3 2 1.34 22 98 prolate 1 EL 0.89

Leo 6.52 5.63 1.16 5.63 28.2 3 0.6 9 88 polygon 1 ER 0.56

Leo 9.57 8.5 1.13 8.5 67.8 3 0.94 15 105 prolate 1 EL 0.98

Leo 4.81 4.61 1.04 4.61 101.5 4 0.88 9 101 spherical 1 ER 0.53

Leo 8.95 5.73 1.56 5.73 44.3 3 0.71 11 126 ovoid 1 EL 0.93

Leo 5.13 4.71 1.09 4.71 19.5 4 0.87 8 90.78 spherical 1 ER 0.86

Leo 9.79 9.26 1.06 9.26 69.3 3 0.83 14 96 spherical 1 EL 0.85

Leo 11.23 9.41 1.19 9.41 97.8 3 1.24 11 86 prolate 1 EL 0.92

Leo 19.42 18.77 1.03 18.77 305.0 2 1.44 11 111 spherical 1 EL 0.67

Leo 7.6 5.68 1.34 5.68 34.6 4 0.83 14 97 ovoid 1 EL 0.99

Leo 10.89 8.71 1.25 8.71 98.8 4 0.8 16 87 ovoid 1 EL 0.99

Leo 10.43 8.09 1.29 8.09 60.3 4 0.92 6 82 prolate 1 ER 0.38

Leo 8.93 8.24 1.08 8.24 60.9 3 0.96 7 90 spherical 1 ER 0.40

Leo 11.82 9.68 1.22 9.68 97.5 4 1.43 19 78 quadrangular 1 EL 0.95

Leo 6.24 5.23 1.19 5.23 27.3 3 0.8 7 81 ovoid 1 ER 0.60

209
Leo 6.44 5.36 1.2 5.36 28.1 3 0.78 6 116 spherical 1 ER 0.74

Leo 5.04 4.34 1.16 4.34 15.2 4 0.72 6 96 polygon 1 ER 0.87

Leo 11.57 11.38 1.02 11.38 100.8 3 0.87 14 121 spherical 1 EL 0.81

Leo 7.44 5.48 1.36 5.48 29.4 3 0.65 9 110 prolate 1 ER 0.63

Leo 14.66 12.68 1.16 12.68 155.9 3 1.33 16 75 prolate 1 EL 0.76

Leo 13.27 9.34 1.42 9.34 94.7 4 0.93 16 96 ovoid 1 EL 0.86

Leo 10.67 8.36 1.28 8.36 71.0 2 1.05 11 81 prolate 1 EL 0.91

Leo 4.77 3.43 1.39 3.43 10.4 3 0.6 6 94 polygon 1 ER 0.56

Leo 5.1 3.67 1.39 3.67 14.6 3 0.6 5 112 ovoid 1 AN 0.48

Leo 6.79 5.43 1.25 5.43 27.0 3 0.75 9 108 spherical 1 ER 0.72

Leo 7.98 7.02 1.14 7.02 44.5 3 0.74 9 96 prolate 1 ER 0.71

Leo 12.85 10.82 1.19 10.82 104.6 3 1.02 12 84 ovoid 1 EL 0.96

Leo 8.05 5.07 1.59 5.07 32.9 4 1.02 12 86 ovoid 1 EL 0.95

Leo 9 7.73 1.16 7.73 54.9 3 1.07 7 98 ovoid 1 AN 0.45

Leo 4.63 3.53 1.31 3.53 14.4 4 0.7 3 91 polygon 1 ER 0.57

Leo 6.31 5.3 1.19 5.3 26.3 2 1 8 117 prolate 1 ER 0.58

Leo 9.47 9.27 1.02 9.27 65.2 2 1.02 15 115 spherical 1 EL 0.73

Leo 9.8 9.66 1.01 9.66 75.7 2 0.92 13 91 spherical 1 EL 0.51

Leo 9.66 9.54 1.01 9.54 77.6 2 1.09 12 107 spherical 1 EL 0.41

Leo 4.3 4.1 1.05 4.1 13.1 4 0.72 7 92.3 polygon 1 ER 0.87

Leo 4.29 4.27 1 4.27 15.4 4 0.65 7 103 polygon 1 ER 0.50

Leo 4.52 4.36 1.04 4.36 17.7 4 0.52 5 110 spherical 1 ER 0.46

Leo 14.83 13.66 1.09 13.66 155.7 3 1.1 19 81 ovoid 1 EL 0.79

Leo 8.58 5.74 1.49 5.74 40.1 4 0.6 12 78 prolate concave-convex 1 EL 0.95

Leo 6.76 5.78 1.17 5.78 39.1 3 0.66 8 94 spherical 1 ER 0.67

Leo 9.02 7.48 1.21 7.48 52.7 4 0.94 10 95 polygon 1 EL 0.53

Leo 5.46 4.38 1.25 4.38 21.2 2 0.5 8 110 spherical 1 ER 0.86

Leo 10.38 6.79 1.53 6.79 66.9 2 0.84 11 78 ovoid 1 EL 0.93

Leo 7.08 6.36 1.11 6.36 40.8 3 0.62 8 119 spherical 1 ER 0.60

Leo 10.21 9.64 1.06 9.64 81.9 1 1.14 10 96 spherical 1 AN 0.42

Leo 24.12 20.27 1.19 20.27 435.5 3 1.8 20 111 prolate 1 EL 0.81

Leo 4.41 3.81 1.16 3.81 14.4 3 0.51 4 110 polygon 1 AN 0.79

Leo 7.76 6.26 1.24 6.26 39.1 3 0.78 16 100 spherical 1 EL 0.58

Leo 7.28 7.17 1.02 7.17 38.1 4 0.7 7 98 polygon 1 ER 0.69

Leo 11.4 10.34 1.1 10.34 96.8 3 1.07 16 89 prolate 1 EL 0.97

Leo 10.4 9.31 1.12 9.31 84.2 4 1.01 13 121 ovoid 1 EL 0.95

Leo 6.99 4.85 1.44 4.85 35.7 4 0.8 9 85 polygon 1 ER 0.37

Leo 13.33 12.72 1.05 12.72 128.6 3 0.92 16 116 spherical 1 EL 0.89

Leo 6.9 5.64 1.22 5.64 35.3 3 0.7 8 100 spherical 1 ER 0.67

Leo 10.67 10.03 1.06 10.03 94.2 2 1 13 105 spherical 1 EL 0.61

Leo 10.89 8.81 1.24 8.81 99.7 3 0.84 10 93 spherical 1 ER 0.44

Rubra 5.03 3.99 1.26 3.99 21.2 4 0.75 9 71 polygon 1 ER 0.52

Rubra 4.32 3.9 1.11 3.9 12.2 3 0.5 5 86 spherical 1 ER 0.44

Rubra 5.14 4.23 1.22 4.23 25.7 3 0.7 6 88 polygon 1 ER 0.86

210
Rubra 11.14 9.56 1.17 9.56 87.5 4 1.3 7 120 ovoid 1 AN 0.42

Rubra 5.65 5.49 1.03 5.49 27.2 3 0.72 6 110 polygon 1 ER 0.66

Rubra 20.7 13.04 1.59 13.04 210.3 4 0.97 19 102 ovoid 1 EL 0.73

Rubra 5.04 3.44 1.47 3.44 13.8 4 0.8 7 72 spherical 1 AN 0.43

Rubra 4.4 3.83 1.15 3.83 12.1 3 0.66 4 114 polygon 1 ER 0.62

Rubra 6.59 3.85 1.71 3.85 19.3 3 0.82 6 70 ovoid 1 EL 0.40

Rubra 9.41 8.54 1.1 8.54 66.2 3 0.87 12 87 spherical 1 EL 0.56

Rubra 13.29 12.04 1.1 12.04 130.2 4 0.75 16 110 ovoid 1 EL 0.86

Rubra 6.04 4.88 1.24 4.88 27.8 4 1.04 7 82.83 polygon 1 AN 0.50

Rubra 11.37 10.36 1.1 10.36 106.3 3 0.6 12 111.16 spherical 1 EL 0.58

Rubra 4.87 4.58 1.06 4.58 17.5 3 0.83 4 100 spherical 1 ER 0.65

Rubra 5.43 5.23 1.04 5.23 23.9 4 0.7 7 110 spherical 1 ER 0.74

Rubra 5.47 4.2 1.3 4.2 15.4 5 0.58 5 99.57 polygon 1 AN 0.47

Rubra 9.63 9.14 1.05 9.14 68.4 3 0.84 7 80 ovoid 1 ER 0.45

Rubra 6.03 5.52 1.09 5.52 26.7 3 0.75 8 56 spherical 1 AF 0.84

Rubra 8.01 6.54 1.22 6.54 39.9 3 0.84 8 86 prolate 1 ER 0.75

Rubra 8.09 7.32 1.11 7.32 49.7 4 0.69 9 95 spherical 1 ER 0.56

Rubra 6.45 4.94 1.31 4.94 26.1 4 0.82 8 91 spherical 1 ER 0.92

Rubra 7.39 4.72 1.57 4.72 32.7 3 0.66 5 89 ovoid 1 ER 0.40

Rubra 6.04 4.64 1.3 4.64 22.0 4 0.85 8 76.18 prolate 1 ER 0.60

Rubra 10.08 7.87 1.28 7.87 63.6 4 0.84 15 86 ovoid 1 EL 1

Rubra 6.29 6.19 1.02 6.19 34.7 3 0.88 5 90 prolate 1 ER 0.74

Rubra 22.19 14.79 1.5 14.79 237.3 2 0.83 15 94 ovoid 1 EL 0.74

Rubra 8.12 6.37 1.27 6.37 45.6 3 0.83 10 99 prolate 1 ER 0.66

Rubra 9.13 8.81 1.04 8.81 65.9 3 0.92 14 104 spherical 1 EL 0.77

Rubra 6.45 4.71 1.37 4.71 23.0 4 0.82 7 92 ovoid 1 ER 0.93

Rubra 10.46 8.45 1.24 8.45 76.9 3 0.84 11 99 ovoid 1 EL 0.95

Rubra 10.67 8.59 1.24 8.59 65.2 4 0.65 7 98 ovoid 1 EL 0.43

Rubra 10.15 8.46 1.2 8.46 83.1 4 1 12 80 ovoid 1 EL 0.97

Rubra 10.55 10.44 1.01 10.44 89.4 4 0.84 16 79.01 spherical 1 EL 0.80

Rubra 12.13 11.81 1.03 11.81 87.4 3 1.17 10 87.43 spherical 1 EL 0.41

Rubra 17.4 12.93 1.35 12.93 180.7 4 0.66 11 107 ovoid 1 EL 0.61

Rubra 12.17 11.04 1.1 11.04 119.0 3 1 17 117 spherical 1 EL 0.95

Rubra 6.42 4.5 1.43 4.5 20.1 4 0.8 7 92 polygon 1 ER 0.71

Rubra 11.8 11.71 1.01 11.71 123.0 4 1.5 12 99 spherical 1 EL 0.55

Rubra 5.74 2.87 2 2.87 20.6 5 2 4 64.16 polygon 1 EL 0.37

Rubra 6.23 6.1 1.02 6.1 35.9 3 0.5 6 101 spherical 1 AF 0.67

Rubra 9.4 7.24 1.3 7.24 56.0 4 0.8 9 107 ovoid 1 ER 0.46

Rubra 28.67 18.76 1.53 18.76 381.8 4 2.4 15 98 ovoid 1 EL 0.71

Rubra 10.63 9.06 1.17 9.06 68.6 4 1.13 11 87 ovoid 1 EL 0.93

Rubra 6.88 5.57 1.24 5.57 31.3 3 0.7 9 110 ovoid 1 ER 0.73

Rubra 10.2 6.57 1.55 6.57 60.5 4 0.92 9 84 ovoid 1 EL 0.36

Rubra 23.9 23.45 1.02 23.45 445.4 3 1.37 15 113 spherical 1 EL 0.82

Rubra 17.74 16.73 1.06 16.73 217.1 3 1.2 17 97 spherical 1 EL 0.89

211
Rubra 5.59 3.62 1.54 3.62 14.0 3 0.8 5 78 polygon 1 ER 0.36

Rubra 16.66 12.46 1.34 12.46 160.8 3 0.72 18 103 ovoid 1 EL 0.75

Rubra 5.49 5.14 1.07 5.14 25.0 4 0.87 10 93 polygon 1 ER 0.67

Rubra 4.45 3.8 1.17 3.8 19.1 4 0.75 6 93 spherical 1 ER 0.92

Rubra 6.25 5.17 1.21 5.17 23.8 3 0.6 9 102 prolate 1 ER 0.72

Rubra 6.6 6.3 1.05 6.3 27.6 3 0.72 6 0.72 spherical 1 AF 0.84

Rubra 17.82 16.19 1.1 16.19 229.1 2 2.3 11 90 spherical 1 EL 0.61

Rubra 13.42 11.74 1.14 11.74 119.4 3 1.11 3 127 prolate 1 EL 0.31

Rubra 14.11 9.72 1.45 9.72 134.1 4 0.7 20 93 prolate 1 EL 0.73

Rubra 12.12 8.93 1.36 8.93 82.4 2 0.9 14 95 prolate 1 EL 0.96

Rubra 12.43 9.34 1.33 9.34 104.8 4 1 14 100 prolate 1 EL 0.98

Rubra 9.32 6.24 1.49 6.24 49.0 3 0.87 15 103 ovoid 1 EL 0.99

Rubra 6.66 5.65 1.18 5.65 33.2 4 0.75 13 95 ovoid 1 EL 0.95

Rubra 11.06 8.74 1.27 8.74 76.1 4 1 15 105 ovoid 1 EL 0.98

Rubra 8.91 8.29 1.07 8.29 58.0 4 0.9 6 111 spherical 1 ER 0.38

Rubra 8.16 6.05 1.35 6.05 35.7 3 0.6 13 95 ovoid 1 EL 0.96

Rubra 6.21 4.21 1.48 4.21 24.9 4 0.65 9 110 polygon 1 ER 0.60

Rubra 14.52 14.44 1.01 14.44 170.7 4 1.23 15 107 ovoid 1 EL 0.77

Rubra 7.85 6.4 1.23 6.4 42.0 4 1 7 87 prolate 1 AN 0.54

Rubra 5.81 5.76 1.01 5.76 27.5 3 0.5 9 115 spherical 1 AN 0.42

Rubra 5.18 3.94 1.31 3.94 16.5 3 0.82 6 92 polygon 1 ER 0.86

Rubra 6.57 6.45 1.02 6.45 33.1 2 0.7 8 90 spherical 1 ER 0.89

Rubra 4 3.5 1.14 3.5 11.9 3 0.5 9 93 spherical 1 ER 0.57

Rubra 6.64 4.69 1.42 4.69 21.4 4 0.7 9 99 polygon 1 ER 0.62

Rubra 8 5.85 1.37 5.85 37.8 4 0.5 11 103 ovoid 1 EL 0.90

Rubra 12.62 11.65 1.08 11.65 107.8 3 0.84 13 122 spherical 1 EL 0.77

Rubra 3.99 3.62 1.1 3.62 13.9 4 0.92 4 99 polygon 1 ER 0.57

Rubra 10.9 10.13 1.08 10.13 91.8 3 0.8 16 86 spherical 1 EL 0.89

Rubra 6.61 6.33 1.04 6.33 35.5 4 0.9 8 97 spherical 1 ER 0.77

Rubra 3.91 3.6 1.09 3.6 12.9 4 0.72 6 98 polygon 1 ER 0.88

Rubra 4.78 4.35 1.1 4.35 19.9 3 0.72 8 101 spherical 1 ER 0.91

Rubra 10.35 8.91 1.16 8.91 81.5 3 0.87 17 88 prolate 1 EL 0.99

Rubra 4.29 3.69 1.16 3.69 13.9 4 0.6 10 90 prolate 1 ER 0.57

Rubra 8.85 5.68 1.56 5.68 33.2 4 0.61 17 80 ovoid 1 EL 0.97

Rubra 6.39 5.84 1.09 5.84 33.0 4 0.82 7 127 polygon 1 ER 0.72

Rubra 6.37 6.01 1.06 6.01 31.7 4 0.93 9 79 spherical 1 ER 0.44

Rubra 4.21 3.19 1.32 3.19 12.8 4 0.6 8 120 ovoid 1 ER 0.62

Rubra 4.61 4.46 1.03 4.46 15.8 4 0.7 4 115 polygon 1 AN 0.50

Rubra 11.29 9.98 1.13 9.98 95.0 5 1.02 20 90 polygon 1 EL 0.98

Rubra 11.28 10.28 1.1 10.28 84.6 3 0.75 15 106 polygon 1 EL 0.96

Rubra 13.8 7.07 1.95 7.07 96.3 5 0.75 11 83 quadrangular 1 SA 0.46

Rubra 9.25 8.91 1.04 8.91 90.2 5 0.75 6 111 ovoid 1 ER 0.39

Rubra 8.3 7.49 1.11 7.49 58.6 3 0.65 17 99 spherical 1 EL 0.75

Rubra 11.23 8.04 1.4 8.04 80.0 3 0.72 9 110 prolate 1 ER 0.39

212
Rubra 3.49 2.39 1.46 2.39 6.2 3 0.6 6 100 polygon 1 ER 0.49

Rubra 5.75 4.55 1.26 4.55 21.5 4 0.65 11 74 polygon 1 EL 0.87

Rubra 12.09 9.11 1.33 9.11 91.6 2 0.65 19 100 ovoid 1 EL 0.94

Rubra 10.04 7.89 1.27 7.89 55.0 4 0.65 7 61 ovoid 1 AF 0.74

Rubra 11.51 9.43 1.22 9.43 90.7 3 0.9 9 119 prolate 1 EL 0.47

Rubra 12.37 11.13 1.11 11.13 124.4 3 1 13 98 ovoid 1 EL 0.96

Rubra 4.73 4.56 1.04 4.56 17.2 4 0.8 6 70 polygon 1 ER 0.49

Rubra 3.72 3.15 1.18 3.15 10.7 3 0.9 7 63 prolate 1 AN 0.71

Rubra 3.86 3.33 1.16 3.33 11.8 3 0.53 6 90 spherical 1 ER 0.62

Rubra 5.89 4.1 1.44 4.1 19.2 4 0.55 9 100 ovoid 1 ER 0.37

Rubra 5.65 4.73 1.19 4.73 26.0 4 0.97 7 100 polygon 1 ER 0.58

Rubra 3.55 2.79 1.27 2.79 8.2 3 0.4 4 85 ovoid 1 AN 0.50

Rubra 5.08 4.12 1.23 4.12 6.5 3 0.51 5 95 ovoid 1 AN 0.47

Noah 3.72 3.39 1.1 3.39 12.2 3 0.6 8 70 polygon 1 ER 0.42

Noah 11.17 9.66 1.16 9.66 83.6 3 0.87 11 97 ovoid 1 EL 0.94

Noah 8.3 7.06 1.18 7.06 50.4 3 0.88 3 99 polygon 1 AN 0.37

Noah 18.72 11.19 1.67 11.19 149.6 3 0.7 20 85 triangular 1 EL 0.67

Noah 7.95 7.11 1.12 7.11 41.5 3 0.8 5 107 spherical 1 ER 0.67

Noah 8.04 8.01 1 8.01 53.8 2 0.92 12 92 spherical 1 ER 0.37

Noah 6.48 5.96 1.09 5.96 28.5 5 1.03 6 0.6 polygon 1 AF 0.79

Noah 14.38 9.27 1.55 9.27 124.1 3 0 0 0 polygon 1 SA 0.92

Noah 2.85 2.8 1.02 2.8 8.6 3 0.43 3 116 polygon 1 AN 0.57

Noah 6.14 5.63 1.09 5.63 32.0 3 0.78 7 101 ovoid 1 ER 0.79

Noah 7.58 7.48 1.01 7.48 36.1 3 0.75 6 89 ovoid 1 ER 0.63

Noah 3.8 3.75 1.01 3.75 12.1 4 0.69 6 111 polygon 1 ER 0.73

Noah 3.01 2.66 1.13 2.66 5.1 3 0.4 5 107 polygon 1 AN 0.37

Noah 12.02 11.19 1.07 11.19 114.4 4 1 16 95 spherical 1 EL 0.95

Noah 9.32 8.7 1.07 8.7 66.7 4 0.9 12 96 spherical 1 EL 0.55

Noah 4.73 4.43 1.07 4.43 15.1 3 0.62 6 80 spherical 1 ER 0.50

Noah 6.32 4.32 1.46 4.32 21.2 4 1 5 74 polygon 1 AN 0.49

Noah 5.39 4.2 1.28 4.2 17.5 4 0.55 5 100 spherical 1 AN 0.55

Noah 7.64 7.52 1.02 7.52 45.0 3 0.7 13 92 spherical 1 ER 0.59

Noah 7.17 6.74 1.06 6.74 46.7 4 0.88 11 56 spherical 1 ER 0.47

Noah 7.37 7.21 1.02 7.21 45.8 4 1 10 106 spherical 1 ER 0.66

Noah 5.12 4.4 1.16 4.4 17.8 9 0.84 11 100 polygon 1 EL 0.86

Noah 7.49 6.83 1.1 6.83 47.9 3 0.78 6 89 polygon 1 ER 0.55

Noah 4.93 4.06 1.21 4.06 18.1 3 0.74 8 99 ovoid 1 ER 0.86

Noah 6.32 6.09 1.04 6.09 31.7 3 0.5 8 102 spherical 1 AF 0.64

Noah 4.74 4.37 1.08 4.37 21.3 4 0.52 4 98 polygon 1 AN 0.77

Noah 7.29 4.45 1.64 4.45 31.7 5 0.94 9 92 polygon 1 ER 0.40

hector 6.32 6.04 1.05 6.04 30.5 4 0.83 7 80 polygon 1 ER 0.60

hector 7.17 4.82 1.49 4.82 30.3 4 0.7 9 101 ovoid 1 ER 0.46

hector 6.59 4.76 1.38 4.76 25.8 4 0.8 7 115 ovoid 1 ER 0.79

hector 5.95 3.79 1.57 3.79 18.5 3 0.72 5 97 prolate 1 ER 0.65

213
hector 11.66 10.63 1.1 10.63 110.0 4 0.97 6 95 spherical 1 EL 0.44

hector 19.32 15.84 1.22 15.84 281.6 1 2.4 15 77.5 prolate 1 EL 0.78

hector 8.24 8.12 1.01 8.12 53.4 2 0.7 12 90 spherical 1 EL 0.41

hector 5.92 5.91 1 5.91 27.9 3 0.72 9 98 spherical 1 ER 0.49

hector 7.65 4.95 1.55 4.95 32.6 4 0.72 4 70 prolate 1 AN 0.42

hector 6.03 3.98 1.52 3.98 27.3 4 0.94 4 77 polygon 1 AN 0.47

hector 4.02 3.5 1.15 3.5 11.4 4 0.83 4 80 polygon 1 ER 0.52

hector 20.32 14.28 1.42 14.28 228.7 3 0.9 30 78 prolate 1 EL 0.81

hector 7.97 7.03 1.13 7.03 43.6 3 1 7 87 ovoid 1 AN 0.52

hector 5.14 4.15 1.24 4.15 16.6 4 0.52 4 100 ovoid 1 AN 0.85

hector 12.98 10.1 1.29 10.1 119.1 4 1.3 8 80.41 ovoid 1 EL 0.47

hector 16.7 13.62 1.23 13.62 198.1 3 1.14 17 71 ovoid 1 EL 0.74

hector 12.6 11.75 1.07 11.75 114.4 3 1 4 120 spherical 1 AN 0.31

hector 12 9.68 1.24 9.68 107.0 4 1 5 100 spherical 1 EL 0.37

hector 19.7 16.43 1.2 16.43 254.7 3 1.44 9 102 ovoid 1 EL 0.47

hector 8.15 7.37 1.11 7.37 53.5 4 1.13 4 84 polygon 1 AN 0.55

hector 12.76 10.63 1.2 10.63 102.6 4 0.87 9 100 spherical 1 EL 0.53

castor 4.52 3.91 1.16 3.91 14.3 3 0.5 5 104 polygon 1 AN 0.51

castor 13.34 11.74 1.14 11.74 112.8 3 0.9 15 116 spherical 1 EL 0.86

castor 6.02 5.47 1.1 5.47 30.0 2 0.65 8 108 spherical 1 ER 0.86

castor 5.63 5.12 1.1 5.12 24.6 2 0.55 7 101 spherical 1 ER 0.80

castor 6.56 5.05 1.3 5.05 31.2 4 0.83 8 59.11 polygon 1 AF 0.86

castor 5.2 4.5 1.16 4.5 20.4 3 0.42 7 108 ovoid 1 ER 0.50

castor 7.31 5.54 1.32 5.54 30.4 4 0.75 11 84.33 polygon 1 EL 0.96

castor 3.77 2.69 1.4 2.69 8.6 3 0.51 3 85 ovoid 1 AN 0.48

castor 6.04 4.3 1.4 4.3 20.7 3 0.4 5 116 prolate 1 AF 0.41

castor 5.95 5.54 1.07 5.54 24.8 2 0.4 7 123 spherical 1 AF 0.39

castor 9.94 6.86 1.45 6.86 48.0 4 0.55 11 90 ovoid 1 EL 0.94

castor 4.53 4.13 1.1 4.13 15.2 4 0.61 6 68 polygon 1 AN 0.39

castor 5.47 5.45 1 5.45 23.0 3 0.46 8 100 spherical 1 AN 0.46

castor 10.56 9.02 1.17 9.02 88.1 4 1.25 6 88.06 spherical 1 AN 0.60

castor 8.29 5.73 1.45 5.73 45.9 5 0.82 5 103 polygon 1 AN 0.35

castor 9.1 6.45 1.41 6.45 47.2 3 1 7 109 polygon 1 ER 0.38

castor 7 6.67 1.05 6.67 32.4 3 0.72 14 68 spherical 1 EL 0.50

castor 6.45 5.6 1.15 5.6 30.0 3 1 8 92.67 spherical 1 ER 0.51

castor 4.54 3.53 1.29 3.53 14.2 3 0.72 3 94 polygon 1 ER 0.58

castor 7.53 5.8 1.3 5.8 28.1 5 0.88 4 77 polygon 1 AN 0.39

castor 2.46 2.35 1.05 2.35 4.7 2 0.42 4 82 polygon 1 ER 0.41

castor 3.58 3.39 1.06 3.39 8.3 4 0.83 2 41 spherical 1 AN 0.79

castor 4.22 3.74 1.13 3.74 14.4 3 0.72 6 71 spherical 1 ER 0.40

castor 5.59 5.45 1.03 5.45 25.6 4 0.6 5 91.25 spherical 1 ER 0.54

castor 13.71 11.85 1.16 11.85 126.0 4 1.25 4 99 spherical 1 SA 0.45

castor 4.74 4.32 1.1 4.32 18.1 2 0.74 8 94.06 spherical 1 ER 0.97

castor 8.76 6.82 1.28 6.82 51.7 3 0.61 17 117.95 ovoid 1 EL 0.96

214
castor 7.54 5.44 1.39 5.44 34.5 3 0.69 6 100 prolate 1 ER 0.50

castor 4.53 3.12 1.45 3.12 13.9 3 0.65 5 78 polygon 2 AN 0.47

castor 6.25 5.96 1.05 5.96 26.5 3 0.72 9 102.89 triangular 2 ER 0.85

castor 6.18 5.34 1.16 5.34 25.9 3 0.46 8 97.41 prolate 1 AF 0.51

castor 6.86 6.25 1.1 6.25 36.7 3 0.88 6 75 ovoid 1 ER 0.46

castor 6.15 4.72 1.3 4.72 23.0 2 0.72 1 65.35 spherical 1 AF 0.73

castor 10.68 8.65 1.23 8.65 65.1 3 0.83 17 88.44 prolate 1 EL 0.99

castor 6.64 4.63 1.43 4.63 26.2 3 0.6 9 90 ovoid 1 ER 0.45

castor 5.22 5 1.04 5 18.8 2 0.75 5 126 spherical 1 ER 0.69

castor 7.18 5.68 1.26 5.68 34.1 4 0.92 7 92 ovoid 1 ER 0.63

castor 5.2 4.72 1.1 4.72 19.9 3 0 0 0 spherical 1 AF 0.90

castor 4.49 3.77 1.19 3.77 12.9 3 0.83 8 89.85 spherical 1 ER 0.91

castor 5.22 5.12 1.02 5.12 21.4 4 0.75 7 85 polygon 1 ER 0.75

castor 7.81 6.09 1.28 6.09 27.7 5 0.83 8 129 polygon 1 ER 0.66

castor 4.06 3.74 1.09 3.74 12.8 4 0.58 5 89 polygon 1 ER 0.53

castor 6.97 5.79 1.2 5.79 33.7 4 0.52 7 88 spherical 1 AF 0.67

castor 5.79 4.71 1.23 4.71 23.1 3 0.61 8 65 spherical 1 AF 0.77

castor 8.93 8.56 1.04 8.56 61.6 3 0.72 9 62 ovoid 1 AF 0.49

castor 4.72 4.24 1.11 4.24 15.9 4 0.97 5 81 polygon 1 AN 0.46

castor 12.18 11.27 1.08 11.27 101.5 3 0.87 5 128.5 ovoid 1 EL 0.38

castor 13.93 11.06 1.26 11.06 128.6 3 1.37 13 55.37 ovoid 1 SA 0.39

castor 5.31 4.96 1.07 4.96 24.1 4 1.01 7 91.91 spherical 1 ER 0.57

castor 4.66 3.99 1.17 3.99 13.9 2 0.52 4 1 prolate 1 AN 0.81

castor 14.06 11.57 1.22 11.57 123.8 2 1.19 14 105 spherical 1 EL 0.78

castor 13.94 11.69 1.19 11.69 119.8 2 0.94 20 75 ovoid 1 EL 0.76

castor 7.29 7.01 1.04 7.01 38.9 3 0.69 7 88 spherical 1 ER 0.66

castor 5.47 4.37 1.25 4.37 17.0 4 0.62 9 84.58 polygon 1 ER 0.62

castor 5.07 4.52 1.12 4.52 18.2 2 0.61 6 85 spherical 1 ER 0.83

castor 5.03 3.54 1.42 3.54 17.9 3 0.75 4 82.1 prolate 1 AN 0.61

castor 10.59 6.86 1.54 6.86 51.3 2 0.83 8 72.34 ovoid 1 ER 0.38

castor 4.1 3.08 1.33 3.08 10.8 3 0.72 6 70 ovoid 1 ER 0.40

castor 5.53 4.82 1.15 4.82 19.3 2 0.61 6 64.66 prolate 1 AF 0.42

castor 9.54 8.72 1.09 8.72 65.2 3 0.94 11 91 prolate 1 EL 0.92

castor 7.05 6.01 1.17 6.01 38.3 3 0.7 9 75 spherical 1 AF 0.50

castor 7.84 5.19 1.51 5.19 34.7 3 0.8 8 96 ovoid 1 ER 0.45

castor 5.33 5.23 1.02 5.23 24.5 4 0.51 7 104 ovoid 1 ER 0.47

castor 4.2 3.9 1.08 3.9 13.9 4 0.52 6 95.19 spherical 1 ER 0.59

castor 9.03 8.81 1.02 8.81 67.7 3 0.83 16 104 spherical 1 EL 0.75

castor 3.9 3.5 1.11 3.5 12.3 3 0.7 7 87 polygon 1 ER 0.87

castor 5.86 4.24 1.38 4.24 19.4 5 0.82 7 90 polygon 1 ER 0.83

castor 6.23 6.06 1.03 6.06 33.8 3 0.72 6 92 ovoid 1 ER 0.75

castor 5.43 4.51 1.2 4.51 19.0 3 0.69 5 95 spherical 1 ER 0.86

bijou 11.52 9.87 1.17 9.87 98.7 3 0.78 13 119.07 prolate 1 EL 0.93

bijou 11.91 9.09 1.31 9.09 80.9 4 0.7 8 130 quadrangular 1 EL 0.48

215
bijou 6.82 6.29 1.08 6.29 34.0 2 0.65 8 96.97 spherical 1 ER 0.87

bijou 8.27 8.23 1 8.23 138.3 5 0.52 7 138.28 polygon 1 AF 0.39

bijou 15.35 9.54 1.61 9.54 125.5 3 0.8 6 125.54 prolate 1 EL 0.40

bijou 9.83 9.73 1.01 9.73 80.8 2 0.65 8 125.06 spherical 1 ER 0.37

bijou 13.78 10.79 1.28 10.79 107.8 3 1.14 15 93.49 spherical 1 EL 0.75

bijou 5.14 4.35 1.18 4.35 19.0 4 0.6 6 123 polygon 1 ER 0.66

bijou 6.02 4.69 1.28 4.69 21.8 3 0.5 4 111.1 polygon 1 AN 0.56

bijou 18.14 13.74 1.32 13.74 225.6 4 1.65 26 97.83 prolate 1 EL 0.79

bijou 10.62 7.8 1.36 7.8 60.0 4 0.93 9 135 polygon 1 EL 0.46

bijou 6.33 5.14 1.23 5.14 36.5 3 0.83 10 104 polygon 1 ER 0.82

bijou 8.81 7.78 1.13 7.78 49.9 3 0.7 9 112.01 spherical 1 ER 0.45

bijou 5.02 4.66 1.08 4.66 21.8 4 0.82 6 98.98 spherical 1 ER 0.90

bijou 6.87 6.66 1.03 6.66 36.7 4 1.03 9 92.28 polygon 1 ER 0.73

bijou 10.58 8.9 1.19 8.9 75.1 3 0.83 12 135.35 spherical 1 EL 0.54

bijou 7.97 6.74 1.18 6.74 43.7 2 0.52 7 107.33 spherical 1 ER 0.43

bijou 19.89 15.17 1.31 15.17 245.6 2 1.33 28 116.95 ovoid 1 EL 0.81

bijou 7.96 5.61 1.42 5.61 41.5 5 1.38 6 94.45 polygon 1 AN 0.50

bijou 8.33 7.18 1.16 7.18 43.9 4 0.9 5 109.88 polygon 1 ER 0.48

bijou 9.55 7.4 1.29 7.4 55.4 4 0.66 6 119.19 ovoid 1 ER 0.39

bijou 6.86 6.07 1.13 6.07 39.1 3 0.6 8 114.04 polygon 1 ER 0.52

bijou 5.91 5.37 1.1 5.37 26.5 5 0.66 7 105.53 spherical 1 ER 0.77

bijou 12.59 11.48 1.1 11.48 108.4 3 1.2 7 107.62 spherical 1 EL 0.42

bijou 9.42 8.42 1.12 8.42 81.4 5 1.47 17 88.21 prolate 1 EL 0.93

bijou 10.09 9.14 1.1 9.14 67.8 3 1.13 11 67.75 prolate 1 EL 0.79

bijou 10.05 9.96 1.01 9.96 85.4 3 0.69 10 84.61 spherical 1 ER 0.39

bijou 10.83 8.3 1.3 8.3 72.4 3 0.6 14 115.75 ovoid 1 EL 0.96

bijou 11.22 10.87 1.03 10.87 100.8 2 0.84 14 123.01 prolate 2 EL 0.84

bijou 19.87 13.21 1.5 13.21 209.9 5 0.97 5 131.4 ovoid 2 SA 0.40

bijou 12.3 9.82 1.25 9.82 96.5 4 0.93 11 108.2 polygon 1 EL 0.91

bijou 11.48 10.24 1.12 10.24 97.3 3 1.01 15 119 ovoid 1 EL 0.96

bijou 7.82 7.2 1.09 7.2 112.3 3 0.55 10 112 spherical 1 ER 0.33

bijou 8.7 7.58 1.15 7.58 49.5 3 0.41 2 134 spherical 1 AF 0.35

bijou 10.92 8.22 1.33 8.22 72.6 4 1.1 8 92.45 prolate 1 AN 0.38

bijou 12.89 11.02 1.17 11.02 119.7 4 1.17 11 122 spherical 1 EL 0.68

bijou 16.28 10.44 1.56 10.44 137.3 3 1.35 6 96.39 ovoid 1 EL 0.38

bijou 10.85 8.37 1.3 8.37 71.2 4 0.92 4 103.66 polygon 1 AN 0.37

bijou 9.11 8.91 1.02 8.91 64.4 2 1.1 12 96.04 spherical 1 EL 0.40

bijou 14.53 14.45 1.01 14.45 172.5 3 1.25 17 114.06 spherical 1 EL 0.86

bijou 9.85 7.73 1.27 7.73 68.6 3 1.23 13 95.02 prolate 1 EL 0.98

bijou 7.69 7.2 1.07 7.2 47.0 4 0.83 14 118.89 polygon 1 EL 0.89

bijou 11.44 9.36 1.22 9.36 81.0 4 0.87 10 114 prolate 1 EL 0.52

bijou 22.53 21.32 1.06 21.32 364.3 2 1.45 15 106.07 ovoid 1 EL 0.84

bijou 7.37 6.46 1.14 6.46 133.6 2 0.43 3 133.55 spherical 1 AN 0.43

bijou 10.69 10.03 1.07 10.03 90.9 3 0.5 16 119.15 spherical 1 EL 0.83

216
bijou 11.42 9.63 1.19 9.63 90.3 4 0.8 6 106 spherical 1 EL 0.46

bijou 8.65 7.37 1.17 7.37 51.5 3 0.92 9 92 spherical 1 ER 0.44

bijou 9.83 9.74 1.01 9.74 74.0 3 0.8 8 95.94 spherical 1 ER 0.44

bijou 11.45 11.13 1.03 11.13 93.3 4 0.72 10 78.76 ovoid 1 EL 0.47

bijou 19.32 17.24 1.12 17.24 256.9 2 0.93 15 116.31 prolate 1 EL 0.80

bijou 11.42 9.29 1.23 9.29 84.1 2 0.4 13 88 ovoid 1 EL 0.90

bijou 6.32 5.93 1.07 5.93 29.7 4 0.78 9 110 polygon 1 ER 0.74

bijou 13.4 11.6 1.16 11.6 136.7 5 1.33 12 104.18 polygon 1 EL 0.80

bijou 14.18 11.63 1.22 11.63 152.9 3 0.9 8 116 polygon 1 EL 0.51

bijou 7.56 6.82 1.11 6.82 43.3 2 0.4 18 82 prolate 1 EL 0.89

bijou 10.31 7.27 1.42 7.27 51.2 2 1.1 8 110 ovoid 1 AN 0.48

bijou 14.13 13.53 1.04 13.53 155.1 5 1.3 15 108.1 polygon 1 EL 0.80

bijou 11.37 10.21 1.11 10.21 89.4 2 0.95 17 83 spherical 1 EL 0.92

bijou 10.79 9.75 1.11 9.75 83.0 3 0.94 9 94 polygon 1 EL 0.51

bijou 7.93 6.62 1.2 6.62 41.9 3 0.97 6 110 polygon 1 ER 0.53

bijou 10.72 9.79 1.09 9.79 78.4 2 0.88 3 66 ovoid 1 AF 0.57

bijou 10.21 9.18 1.11 9.18 73.6 2 0.75 10 90 spherical 1 ER 0.46

Goma 7.82 6.04 1.29 6.04 37.9 2 0 0 0 spherical 1 AF 0.99

Goma 6.77 6.24 1.08 6.24 33.2 2 0.51 10 100 spherical 1 ER 0.59

Goma 11.87 6.68 1.78 6.68 54.7 0 0 0 0 spherical 1 SA 0.74

Goma 20.04 18.05 1.11 18.05 303.1 2 1.42 19 91 spherical 1 EL 0.85

Goma 11.59 8.57 1.35 8.57 75.2 3 1.2 13 103 ovoid 1 EL 0.95

Goma 6.65 5.86 1.13 5.86 30.8 5 0 0 0 quadrangular 1 AF 0.98

Goma 8.44 6.95 1.21 6.95 49.3 4 1.01 14 95.03 ovoid 1 EL 1

Goma 5.43 4.51 1.2 4.51 19.0 4 0 0 0 polygon 1 AN 0.51

Goma 23.45 21.32 1.1 21.32 398.3 3 1.43 20 111 spherical 1 EL 0.82

Goma 10.35 8.7 1.19 8.7 66.2 3 1.74 7 66.17 spherical 1 EL 0.40

Goma 17.63 14.66 1.2 14.66 220.2 2 1.62 22 124.85 spherical 1 EL 0.84

Goma 17.3 13.53 1.28 13.53 215.0 3 1.44 16 101.93 prolate 1 EL 0.81

Goma 15.01 13.01 1.15 13.01 143.6 1 1.04 8 123.41 spherical 1 EL 0.49

Goma 8.7 7.92 1.1 7.92 57.5 3 1.17 10 57.51 spherical 1 AF 0.44

Goma 6.58 4.82 1.37 4.82 25.0 4 0.7 7 24.95 prolate 1 AF 0.77

Goma 6.38 5.8 1.1 5.8 38.2 3 0 0 38.2 quadrangular 1 AF 0.98

Goma 19.45 15.59 1.25 15.59 243.7 4 1.33 16 243.69 spherical 1 EL 0.81

Goma 6.72 5.87 1.14 5.87 36.0 2 0.7 6 36.02 spherical 1 AF 0.94

Goma 17.93 12.92 1.39 12.92 191.2 2 1.27 11 80.54 ovoid 1 EL 0.67

Goma 16.05 13.33 1.2 13.33 167.2 4 1.1 8 94 ovoid 1 EL 0.47

Goma 10.73 9.49 1.13 9.49 76.3 5 0.75 8 117.27 spherical 1 ER 0.38

Goma 11.77 8.81 1.34 8.81 88.9 5 0.7 5 11.97 polygon 1 AF 0.54

Goma 15.02 12.36 1.22 12.36 155.1 3 1.3 13 100.25 prolate 1 EL 0.75

Goma 20.3 18.26 1.11 18.26 303.7 3 1.69 24 89.34 spherical 1 EL 0.81

Goma 15 11.29 1.33 11.29 133.1 2 1.14 11 91.53 prolate 1 EL 0.68

Goma 17.58 16.61 1.06 16.61 118.6 4 1.23 11 118.55 spherical 1 EL 0.64

Goma 11.38 8.86 1.28 8.86 88.9 3 1.11 5 105 spherical 1 AN 0.42

217
Goma 9.9 8.76 1.13 8.76 68.6 4 0.88 18 110 ovoid 1 EL 0.98

Goma 10.08 8.46 1.19 8.46 56.6 4 1.02 15 118.21 prolate 2 EL 0.98

Goma 10.26 6.6 1.55 6.6 66.8 4 1.02 15 103 spherical 2 EL 0.72

Goma 19.89 16.12 1.23 16.12 270.4 4 1.96 18 92.42 polygon 1 EL 0.80

Goma 10.4 8.47 1.23 8.47 77.0 2 0.93 19 84.4 ovoid 1 EL 0.99

Goma 19.39 14.23 1.36 14.23 220.1 4 1.33 24 112.26 ovoid 1 EL 0.78

Goma 19.03 18.8 1.01 18.8 167.1 3 1.74 13 267.06 spherical 1 EL 0.78

Goma 22.73 20.08 1.13 20.08 360.2 3 1.33 11 118.62 spherical 1 EL 0.59

Goma 15.72 14.09 1.12 14.09 177.5 4 1.43 8 177.49 spherical 1 EL 0.47

Goma 17.72 15.08 1.18 15.08 198.7 4 0.94 13 110.35 spherical 1 EL 0.73

Goma 7.47 6.86 1.09 6.86 43.5 4 0.6 15 113.87 spherical 1 EL 0.53

Goma 13.92 13.01 1.07 13.01 159.4 3 1.33 11 120.31 spherical 1 EL 0.62

Goma 14.35 13.62 1.05 14.35 164.4 4 1.33 12 109 spherical 1 EL 0.65

Goma 9.23 8.5 1.09 8.5 65.7 2 0.6 9 106 spherical 1 ER 0.44

Goma 7.3 6.04 1.21 6.04 34.5 5 0.55 7 102 ovoid 1 AF 0.58

Goma 8.33 6.87 1.21 6.87 50.3 4 0.74 13 106.54 spherical 1 EL 0.48

Goma 10.11 7.91 1.28 7.91 69.5 3 0.52 4 114.23 ovoid 1 AN 0.39

Goma 9.85 8.27 1.19 8.27 65.9 4 1.01 9 105.36 polygon 1 EL 0.43

Goma 21.3 16.08 1.32 16.08 263.6 5 1.07 22 110.99 polygon 1 EL 0.79

Goma 9.41 7.37 1.28 7.37 60.6 5 0 0 0 polygon 1 AF 0.99

Goma 16.73 13.74 1.22 13.74 186.0 3 0.75 4 99 ovoid 1 SA 0.40

Goma 18.85 13.74 1.37 13.74 190.9 5 1.07 17 99 ovoid 1 EL 0.78

Goma 18.98 9.32 2.04 9.32 144.8 5 0 0 0 angular point 1 SA 1

Goma 14.96 14.44 1.04 14.44 172.8 1 1.13 18 91 spherical 1 EL 0.88

Goma 16.21 13.09 1.24 13.09 165.4 4 1.25 16 101 ovoid 1 EL 0.77

Goma 11.53 9.88 1.17 9.88 82.8 3 0.88 5 113 spherical 1 EL 0.38

Goma 10.38 9.82 1.06 9.82 82.5 5 0.92 7 106 polygon 1 EL 0.43

Goma 10.71 10.59 1.01 10.59 100.2 3 0.97 8 99.67 spherical 1 ER 0.43

Goma 12.15 10.2 1.19 10.2 93.2 3 1.03 4 104.17 ovoid 1 EL 0.32

Goma 9.06 8.11 1.12 8.11 58.8 3 0.88 10 103.67 ovoid 1 ER 0.44

Goma 11.26 8.65 1.3 8.65 79.3 4 1.31 10 103.89 spherical 1 AN 0.45

Goma 13.34 13.19 1.01 13.19 144.8 2 1.11 8 92.68 spherical 1 EL 0.47

Goma 8.13 7.68 1.06 7.68 49.8 3 0 0 0 polygon 1 AF 0.90

Goma 18.91 16.33 1.16 16.33 265.9 2 0.9 17 99 ovoid 1 EL 0.82

Goma 9.55 8.57 1.11 8.57 71.1 4 1.38 3 75 ovoid 1 AN 0.59

Goma 8.87 6.77 1.31 6.77 47.4 5 0 0 0 ovoid 1 AF 1

Goma 8.66 7.28 1.19 7.28 54.3 4 0.6 5 100 spherical 1 EL 0.31

Goma 7.24 6.23 1.16 6.23 35.4 3 0.5 4 87.1 spherical 1 AF 0.51

Goma 10.4 7.27 1.43 7.27 64.5 3 1 5 104 prolate 1 AN 0.42

Zerlina 7.95 7.14 1.11 7.14 38.2 3 0.75 8 111 spherical 1 ER 0.74

Zerlina 12.24 12.04 1.02 12.04 124.2 3 0.84 6 98 spherical 1 EL 0.48

Zerlina 8.98 8.26 1.09 8.26 69.8 4 1.02 9 89 polygon 1 EL 0.43

Zerlina 6.31 5.74 1.1 5.74 27.7 4 1.23 8 90 spherical 1 AN 0.65

Zerlina 3.99 3.58 1.11 3.58 12.1 4 0.8 5 92 spherical 1 ER 0.88

218
Zerlina 6.69 5.07 1.32 5.07 34.8 3 0.97 9 78 ovoid 1 AN 0.59

Zerlina 20.79 16.53 1.26 16.53 267.0 4 1.23 15 118.71 prolate 1 EL 0.76

Zerlina 4.11 3.69 1.11 3.69 12.0 3 0.6 7 83 spherical 1 ER 0.50

Zerlina 22.42 21.32 1.05 21.32 386.9 3 1.75 17 93 spherical 1 EL 0.89

Zerlina 16.93 15.06 1.12 15.06 243.0 3 0.83 18 101 spherical 1 EL 0.81

Zerlina 12.88 9.43 1.37 9.43 100.0 3 0.97 8 116 prolate 1 EL 0.44

Zerlina 12.72 11.34 1.12 11.34 98.5 3 0.94 12 98 prolate 1 EL 0.95

Zerlina 8.64 6.55 1.32 6.55 37.9 4 1.23 6 86 polygon 1 AN 0.63

Zerlina 14.65 13.51 1.08 13.51 175.5 4 1.17 14 82 spherical 1 EL 0.81

Zerlina 7.23 6.99 1.03 6.99 42.1 4 1.11 7 88 spherical 1 AN 0.53

Zerlina 15.67 15.08 1.04 15.08 196.2 4 0.97 14 114 spherical 1 EL 0.82

Zerlina 17.38 14.84 1.17 14.84 200.9 3 1.25 19 106 spherical 1 EL 0.82

Zerlina 3.88 3.42 1.13 3.42 12.0 3 0.72 3 91 polygon 1 ER 0.72

Zerlina 10.79 8.97 1.2 8.97 78.9 3 1.11 14 116 prolate 1 EL 0.98

Zerlina 12.26 11.27 1.09 11.27 118.3 2 0.84 16 109 spherical 1 EL 0.95

Zerlina 9.43 8.1 1.16 8.1 61.7 4 0.97 11 103 ovoid 1 EL 0.93

Zerlina 8.12 5.83 1.39 5.83 37.3 3 0.51 7 120 prolate 1 AF 0.45

Zerlina 6.56 5.84 1.12 5.84 32.9 3 0.55 10 99 spherical 1 AF 0.39

Zerlina 15.96 12.03 1.33 12.03 167.6 3 1.37 17 97 ovoid 1 EL 0.77

Zerlina 17.85 17.92 1 17.92 257.0 3 1.5 17 84 ovoid 1 EL 0.83

Zerlina 6.9 5.71 1.21 5.71 32.3 3 0.83 6 98 spherical 1 ER 0.63

Zerlina 10.88 10.17 1.07 10.17 85.0 4 0.92 8 88 polygon 1 EL 0.49

Zerlina 9.69 6.64 1.46 6.64 112.4 4 1.31 10 112 ovoid 1 AN 0.54

Zerlina 11.53 9.26 1.25 9.26 91.0 4 0.8 8 121 ovoid 1 EL 0.48

Zerlina 6.38 4.43 1.44 4.43 26.0 4 0.62 9 100 polygon 1 ER 0.53

Zerlina 9.82 7 1.4 7 58.0 4 1.11 7 79 polygon 1 AN 0.53

Zerlina 5.49 4.32 1.27 4.32 20.3 4 0.78 5 93 polygon 1 ER 0.82

Zerlina 9.52 8.4 1.13 8.4 54.7 4 0.7 11 89 ovoid 1 EL 0.96

Zerlina 4.3 4.32 1 4.32 13.8 4 0.52 6 115 polygon 1 AN 0.59

Zerlina 12.24 11.42 1.07 11.42 110.6 2 1.17 11 89 prolate 1 EL 0.90

Zerlina 9.52 5.95 1.6 5.95 45.9 4 0.72 9 99.74 ovoid 1 EL 0.43

Zerlina 7.9 6.03 1.31 6.03 40.5 4 0.8 5 93.7 ovoid 1 ER 0.64

Zerlina 4.93 3.98 1.24 3.98 15.6 3 0.5 4 90 polygon 1 AN 0.86

Zerlina 17.35 12.7 1.37 12.7 167.6 3 1.17 18 93 prolate 1 EL 0.80

Zerlina 9.81 7.9 1.24 7.9 57.1 3 1.07 17 95 ovoid 1 EL 1

Zerlina 10.51 9.36 1.12 9.36 65.6 3 1.23 7 107 spherical 1 AN 0.45

Zerlina 15.19 13.32 1.14 13.32 159.0 3 1.28 14 104 spherical 1 EL 0.76

Zerlina 9.83 7.39 1.33 7.39 56.2 4 1.44 2 86 spherical 1 AN 0.67

Zerlina 6.71 5.91 1.14 5.91 32.5 4 0.83 11 88 polygon 1 EL 0.91

Zerlina 4.92 4.69 1.05 4.69 20.0 3 0.83 6 81 spherical 1 ER 0.53

Zerlina 11.8 10.66 1.11 10.66 101.4 3 0.97 12 102 spherical 1 EL 0.68

Zerlina 6.36 4.1 1.55 4.1 24.9 3 0.61 5 107 ovoid 1 ER 0.52

Zerlina 9.21 8.92 1.03 8.92 71.6 2 0.92 13 95.76 spherical 1 EL 0.52

Zerlina 9.71 8.03 1.21 8.03 75.8 3 1.17 11 81 prolate 1 EL 0.91

219
Zerlina 7.42 5.5 1.35 5.5 34.5 3 0.75 10 75 prolate 1 ER 0.42

Zerlina 12.39 11.18 1.11 11.18 112.9 2 0.83 19 110 spherical 1 EL 0.94

Zerlina 11.95 7.82 1.53 7.82 76.6 3 1.09 14 108 ovoid 1 EL 0.94

Zerlina 8.39 8.33 1.01 8.33 60.9 4 1.1 6 115 polygon 1 EL 0.47

Zerlina 13.4 12.85 1.04 12.85 139.6 3 1.13 17 103 prolate concave-convex 1 EL 0.89

Zerlina 13.58 12.68 1.07 12.68 130.8 3 1.11 15 105 spherical 1 EL 0.82

Zerlina 15.75 15.57 1.01 15.57 178.5 3 1.02 20 115 ovoid 1 EL 0.82

Zerlina 16.36 12.5 1.31 12.5 163.0 3 1.7 13 86 ovoid 1 EL 0.73

Zerlina 8.81 7.71 1.14 7.71 63.3 4 0.97 8 87 polygon 1 EL 0.43

Zerlina 10.72 10.61 1.01 10.61 93.8 3 1.09 18 94 spherical 1 EL 0.76

Zerlina 11.15 10.58 1.05 10.58 104.7 3 1.14 16 96 spherical 1 EL 0.85

Zerlina 7.03 6.38 1.1 6.38 36.1 4 0.88 7 87 prolate 1 ER 0.72

Zerlina 11.13 9.49 1.17 9.49 86.9 4 0.88 17 90.79 ovoid 1 EL 0.98

Zerlina 7.17 6.86 1.05 6.86 44.2 4 0.84 9 111 spherical 1 ER 0.72

Zerlina 9.52 8.6 1.11 8.6 69.8 3 0.94 10 76 spherical 1 ER 0.37

Zerlina 16.37 11.61 1.41 11.61 146.2 3 1.23 19 101 ovoid 1 EL 0.77

Zerlina 15.09 12.89 1.17 12.89 145.6 4 1.3 11 106 ovoid 1 EL 0.63

Zerlina 9.34 7.45 1.25 7.45 55.1 5 1.33 7 97 polygon 1 EL 0.42

Zerlina 11.98 9.72 1.23 9.72 108.2 4 1.01 9 119 ovoid 1 EL 0.46

Zerlina 11.37 9.22 1.23 9.22 90.2 3 1.23 13 107 spherical 1 EL 0.69

Zerlina 21.3 18.12 1.18 18.12 311.0 2 1.17 18 101 ovoid 1 EL 0.82

Zerlina 7.9 5.51 1.43 5.51 37.3 2 0.75 12 89 ovoid 1 EL 0.95

Zerlina 12.02 7.25 1.66 7.25 61.7 3 0.83 7 106 ovoid 1 EL 0.47

Zerlina 17.2 15.46 1.11 15.46 219.0 3 1.36 18 106 spherical 1 EL 0.82

Zerlina 17.47 13.35 1.31 13.35 165.4 3 1.37 17 92 ovoid 1 EL 0.79

Zerlina 5.73 5.43 1.06 5.43 26.0 3 0.6 4 75 prolate concave-convex 1 AN 0.43

Zerlina 3.79 3.58 1.06 3.58 11.0 3 0.42 3 108 polygon 1 AN 0.69

Dorry 6.41 4.61 1.39 4.61 21.0 2 0.69 13 74 prolate 1 EL 0.95

Dorry 7.77 5.72 1.36 5.72 40.2 3 0.9 13 77 prolate 1 EL 0.96

Dorry 7.69 7.07 1.09 7.07 43.8 4 0.55 15 97 spherical 1 EL 0.54

Dorry 7.17 6.76 1.06 6.76 36.7 3 1.09 11 89 spherical 1 ER 0.51

Dorry 6.04 5.77 1.05 5.77 25.7 3 0.65 8 93 spherical 1 ER 0.77

Dorry 7.03 5.29 1.33 5.29 31.6 4 0.72 7 94 ovoid 1 ER 0.71

Dorry 10.49 9.32 1.13 9.32 67.0 3 0.83 12 108 spherical 1 EL 0.55

Dorry 5.98 5.74 1.04 5.74 33.6 3 0.65 15 82 prolate 1 EL 0.74

Dorry 5.66 5 1.13 5 26.5 2 0.83 8 92 spherical 1 ER 0.95

Dorry 6.99 5.55 1.26 5.55 31.0 3 0.83 8 105 spherical 1 ER 0.70

Dorry 9.88 8.26 1.2 8.26 69.4 4 1.1 13 97 ovoid 1 EL 0.98

Dorry 8.73 8.13 1.07 8.13 56.6 4 0.88 12 105 spherical 1 EL 0.54

Dorry 6.17 4.95 1.25 4.95 22.4 3 0.74 7 90 prolate 1 ER 0.98

Dorry 10.09 8.34 1.21 8.34 68.0 4 0.78 16 90 spherical 1 EL 0.87

Dorry 9.67 7.47 1.29 7.47 62.1 4 1.35 14 76 spherical 1 EL 0.80

Dorry 9.96 9.27 1.07 9.27 71.0 3 0.8 10 111 spherical 1 ER 0.46

220
Dorry 8.91 6.15 1.45 6.15 43.6 4 0.88 9 89 ovoid 1 ER 0.39

Dorry 7.52 5.14 1.46 5.14 31.7 5 0.52 6 107 prolate 1 AF 0.45

Dorry 5.41 4.73 1.14 4.73 21.9 2 0.87 8 85 spherical 1 ER 0.78

Dorry 7.78 5.8 1.34 5.8 45.5 5 1.13 9 76 ovoid 1 AN 0.84

Dorry 10.78 7.32 1.47 7.32 55.0 3 0.62 15 98 ovoid 1 EL 0.98

Dorry 10.03 9.63 1.04 9.63 75.0 3 0.9 13 79 spherical 1 EL 0.60

Dorry 5.62 3.71 1.51 3.71 21.9 3 0.6 7 100 prolate 1 ER 0.51

Dorry 5.53 5.43 1.02 5.43 22.3 3 0.83 9 62 spherical 1 AF 0.54

Dorry 12 10.81 1.11 10.81 124.9 3 0.94 15 91 spherical 1 EL 0.94

Dorry 5.33 4.4 1.21 4.4 20.3 4 0.74 7 89 prolate 1 ER 0.93

Dorry 10.86 8.18 1.33 8.18 65.9 4 1.14 9 96 ovoid 1 AN 0.40

Dorry 4.87 4.36 1.12 4.36 21.5 3 0.83 6 83 spherical 1 ER 0.57

Dorry 4.92 3.92 1.26 3.92 16.6 4 0.51 7 95 ovoid 1 ER 0.49

Dorry 9.5 8.36 1.14 8.36 55.7 4 0.61 9 118 ovoid 1 EL 0.46

Dorry 6.29 3.52 1.79 3.52 22.3 5 1.19 9 89 prolate 1 AN 0.55

Dorry 5.77 4.54 1.27 4.54 24.1 3 0.74 6 82 spherical 1 ER 0.53

Dorry 24.69 21.2 1.16 21.2 353.3 1 2.7 19 79 spherical 1 EL 0.82

Dorry 10.65 9.65 1.1 9.65 86.3 3 0.82 11 106 spherical 1 EL 0.51

Dorry 10.31 8.01 1.29 8.01 61.6 3 0.87 16 86 ovoid 1 EL 1

Dorry 6.14 5.45 1.13 5.45 26.2 4 0.92 9 94 spherical 1 ER 0.84

Dorry 5.43 5.22 1.04 5.22 34.4 3 0.84 10 91 spherical 1 ER 0.79

Dorry 6.45 5.63 1.15 5.63 31.8 3 0.66 10 95 spherical 1 ER 0.76

Dorry 9.85 6.85 1.44 6.85 46.0 5 1.1 8 94 triangular 1 AN 0.56

Dorry 10.85 10.44 1.04 10.44 87.9 3 0.93 9 100 spherical 1 ER 0.45

Dorry 7.78 6.27 1.24 6.27 38.8 3 0.8 7 89 spherical 1 ER 0.70

Dorry 11.64 10.42 1.12 10.42 105.9 4 1.14 13 113 spherical 1 EL 0.77

Dorry 10.36 8.8 1.18 8.8 74.0 3 0.78 12 111 prolate 1 EL 0.95

Dorry 4.83 4.37 1.11 4.37 16.3 3 0.65 5 100 spherical 1 ER 0.80

Dorry 8.3 5.63 1.47 5.63 34.8 4 0.82 8 88 ovoid 1 ER 0.36

Dorry 7.83 6.25 1.25 6.25 39.2 4 0.72 10 93 ovoid 1 ER 0.71

Dorry 3.53 2.42 1.46 2.42 7.8 3 0.52 5 93 prolate 1 ER 0.37

Dorry 4.61 4.61 1 4.61 13.5 5 0.51 8 75 spherical 1 AN 0.43

Dorry 3.67 3.02 1.22 3.02 12.2 4 0.55 6 111 spherical 1 ER 0.54

Dorry 5.94 4.49 1.32 4.49 25.9 3 0.75 11 102 spherical 1 ER 0.59

Dorry 10.15 7.19 1.41 7.19 65.1 3 0.74 13 89 prolate 1 EL 0.99

Dorry 7.07 5.73 1.23 5.73 28.1 2 0.75 14 90 prolate 1 EL 0.98

Dorry 6.86 5.86 1.17 5.86 31.2 5 0.94 12 87 ovoid 1 EL 0.92

Dorry 7.69 6.17 1.25 6.17 42.8 3 0.78 12 102 prolate 1 EL 0.92

Dorry 5.46 4.29 1.27 4.29 20.3 4 0.61 7 106 ovoid 1 ER 0.71

Dorry 5.42 4.92 1.1 4.92 22.5 3 0.82 3 105 spherical 1 ER 0.68

Dorry 15.05 14.44 1.04 14.44 175.6 2 0.78 12 105 spherical 1 EL 0.66

Dorry 6.74 5.99 1.13 5.99 29.4 3 0.88 7 90 spherical 1 ER 0.70

Dorry 4.41 3.71 1.19 3.71 14.1 3 0.51 8 85 spherical 1 ER 0.55

Dorry 7.16 6.04 1.19 6.04 50.1 3 0.66 10 107 spherical 1 ER 0.64

221
Dorry 7.07 6.56 1.08 6.56 42.3 2 0.93 11 81 spherical 1 ER 0.55

Dorry 5.33 4.81 1.11 4.81 19.0 3 0.84 9 93.33 spherical 1 ER 0.85

Dorry 7.66 5.68 1.35 5.68 36.6 3 1.05 11 83 spherical 1 ER 0.45

Dorry 7.28 5.53 1.32 5.53 30.4 3 0.97 14 103 prolate 1 EL 0.97

Dorry 5.94 5.33 1.11 5.33 29.7 3 0.87 11 108 spherical 1 ER 0.59

Dorry 7.17 4.98 1.44 4.98 33.9 2 0.72 11 87 prolate 1 EL 0.94

Dorry 9.17 8.91 1.03 8.91 67.1 2 0.87 13 120 spherical 1 EL 0.52

Dorry 10.46 10.03 1.04 10.03 77.9 4 0.97 19 99 spherical 1 EL 0.78

Dorry 6.79 5.4 1.26 5.4 27.1 3 0.65 13 91 ovoid 1 EL 0.96

Dorry 18.76 14.9 1.26 14.9 236.0 1 1.88 32 60 ovoid 1 EL 0.50

Dorry 19.78 17.67 1.12 17.67 271.0 3 1.48 11 99 spherical 1 EL 0.59

Dorry 5.63 5.33 1.06 5.33 22.6 3 0.61 7 97 spherical 1 ER 0.71

Dorry 6.81 5.58 1.22 5.58 28.3 3 0.51 9 106 prolate 1 AF 0.44

Dorry 7.37 5.73 1.29 5.73 34.9 4 0.88 16 97 spherical 1 EL 0.59

Dorry 11.81 10.41 1.13 10.41 100.0 5 1.13 7 90 spherical 1 EL 0.42

Dorry 5.26 4.1 1.28 4.1 20.0 3 0.75 6 91 polygon 1 ER 0.88

Dorry 6.37 5.07 1.26 5.07 27.8 5 0.72 12 95 polygon 1 EL 0.93

Dorry 5.09 4.67 1.09 4.67 19.6 3 0.62 7 80 ovoid 1 ER 0.55

Dorry 5.61 5.29 1.06 5.29 21.7 3 0.61 5 98 polygon 1 ER 0.68

Dorry 8.5 8 1.06 8 54.1 2 1.01 11 83 spherical 1 EL 0.48

Dorry 7.48 7.18 1.04 7.18 35.6 3 0.93 7 95 spherical 1 ER 0.66

Dorry 8.16 6.47 1.26 6.47 38.8 2 1.25 9 100 spherical 1 AN 0.61

Dorry 6.12 4.6 1.33 4.6 25.8 3 1 9 70 prolate 1 AN 0.62

Dorry 3.7 2.97 1.25 2.97 7.0 3 0.61 4 91 spherical 1 ER 0.65

Dorry 8.55 8.13 1.05 8.13 59.4 3 0.8 13 92 ovoid 1 EL 0.92

Dorry 5.37 3.3 1.63 3.3 13.1 5 0.72 7 90 polygon 1 ER 0.61

Venus 5.07 4.49 1.13 4.49 17.2 3 0.55 6 99 spherical 1 ER 0.57

Venus 7.42 2.73 2.72 2.73 19.0 5 0 0 0 angular point 1 EL 0.43

Venus 5.23 4.3 1.22 4.3 19.2 3 0.61 9 70 spherical 1 ER 0.33

Venus 4.56 3.33 1.37 3.33 11.7 3 0.75 7 70 spherical 1 ER 0.36

Venus 6.16 4.25 1.45 4.25 22.4 3 0.8 7 88 spherical 1 ER 0.61

Venus 5.07 3.79 1.34 3.79 16.4 3 0.6 7 84 prolate 1 ER 0.55

Venus 5.77 4.22 1.37 4.22 19.5 4 0.83 6 101 spherical 1 ER 0.84

Venus 8.32 7.31 1.14 7.31 50.9 3 0.75 10 100 spherical 1 ER 0.43

Venus 6.54 4.74 1.38 4.74 26.5 5 0.88 6 94 prolate 1 ER 0.77

Venus 6.83 4.5 1.52 4.5 34.1 4 0.63 8 85 prolate 1 ER 0.32

Venus 4.3 3.28 1.31 3.28 12.9 4 0.51 5 100 polygon 1 AN 0.49

Venus 5.54 4.81 1.15 4.81 18.3 3 0.78 8 85 spherical 1 ER 0.74

Venus 7.07 5.63 1.26 5.63 34.5 3 1 7 88 spherical 1 AN 0.64

Venus 3.72 3.66 1.02 3.66 10.5 4 0.72 5 85 polygon 1 ER 0.73

Venus 6.59 4.99 1.32 4.99 25.0 3 0.72 8 88 polygon 1 ER 0.85

Venus 7.08 5.12 1.38 5.12 28.4 4 0.62 5 98 polygon 1 ER 0.47

Venus 7.16 5.45 1.31 5.45 36.1 3 0.69 10 78 prolate 1 ER 0.42

Venus 2.82 2.53 1.11 2.82 5.9 4 0.75 4 101 polygon 1 ER 0.68

222
Venus 5.45 4.95 1.1 5.45 27.1 4 0.8 9 89 ovoid 1 ER 0.81

Venus 5.33 3.49 1.53 3.49 14.2 5 0.75 7 95 ovoid 1 ER 0.61

Venus 12.7 7.89 1.61 7.89 72.0 2 0.8 12 107 ovoid 1 EL 0.87

Venus 6.14 4.64 1.32 4.64 30.6 5 1.24 5 78.06 polygon 1 AN 0.65

Venus 9.24 7.7 1.2 7.7 55.6 4 0.87 8 55.6 spherical 1 AF 0.85

Venus 9.27 8.47 1.09 8.47 67.0 5 1.64 4 68 polygon 1 EL 0.43

Venus 5.36 4.13 1.3 4.13 18.4 3 0.7 7 87 spherical 1 ER 0.80

Venus 4.71 3.69 1.28 3.69 12.0 4 0.6 5 79 polygon 1 ER 0.45

Venus 7.58 5.53 1.37 5.53 31.5 3 0.4 8 110 ovoid 1 AF 0.59

Venus 8.31 3.18 2.61 3.18 8.3 4 0.75 4 75 polygon 1 EL 0.56

Venus 4.6 4 1.15 4 16.1 2 0.6 7 100 prolate 1 ER 0.87

Venus 7.62 5.73 1.33 5.73 34.3 3 0.69 6 80 ovoid 1 ER 0.38

Venus 10.35 7.95 1.30 7.95 63.9 4 0.87 9 99 spherical 1 ER 0.44

Venus 4.8 4.56 1.05 4.56 17.9 4 0.62 5 77 polygon 1 ER 0.48

Venus 5.9 3.41 1.73 3.41 16.5 5 0.62 8 82 polygon 1 EL 0.46

Venus 5.14 4.22 1.22 4.22 15.7 3 0.72 7 97 ovoid 1 ER 0.83

Venus 6.53 4.92 1.33 4.92 25.1 3 0.62 7 82.14 prolate 1 ER 0.53

Venus 12.32 10.32 1.19 10.32 97.7 3 1 9 91 ovoid 1 EL 0.46

Venus 4.51 4.1 1.1 4.1 16.6 4 0.58 5 95 polygon 1 ER 0.51

Venus 5.92 4.72 1.26 4.72 24.2 4 0.65 9 102 polygon 1 ER 0.84

Venus 7.8 5.74 1.36 5.74 31.7 5 1.44 4 61 polygon 1 AN 0.41

Venus 7.69 6.36 1.21 6.36 35.3 3 0.78 8 98 spherical 1 ER 0.75

Venus 5.8 5.08 1.141 5.08 25.3 3 0.66 8 92 spherical 1 ER 0.89

Venus 4.1 3.99 1.03 3.99 13.8 3 0.6 7 88 polygon 1 ER 0.70

Venus 4.51 4.4 1.03 4.4 13.5 3 0.61 7 91 polygon 1 ER 0.75

Venus 6.86 5.84 1.17 5.84 32.1 3 0.72 9 96 polygon 1 ER 0.75

Venus 3.1 2.78 1.12 2.78 9.0 4 0.74 5 80 polygon 1 ER 0.60

Venus 4.34 3.72 1.17 3.72 14.2 4 0.84 7 89 polygon 1 ER 0.90

Venus 6.15 5.84 1.05 5.84 30.8 5 0.97 7 88 polygon 1 ER 0.50

fanny 5.24 4.12 1.27 4.12 15.3 3 0.82 11 95 spherical 1 ER 0.58

fanny 7.11 6.6 1.08 6.6 39.2 3 0.7 9 119 spherical 1 ER 0.72

fanny 9.56 8 1.2 8 66.3 3 1.15 10 65 spherical 1 EL 0.32

fanny 8.61 6.65 1.29 6.65 41.9 4 0.85 15 98 spherical 1 EL 0.67

fanny 10.08 8.94 1.13 8.94 73.2 3 1 15 92 spherical 1 EL 0.86

fanny 6.97 6.45 1.08 6.45 38.7 3 0.72 8 80 spherical 1 ER 0.48

fanny 10.46 7.24 1.44 7.24 67.5 3 0.85 14 80 ovoid 1 EL 1

fanny 9.83 6.65 1.48 6.65 55.1 4 0.61 7 112 ovoid 1 EL 0.31

fanny 5.8 5.36 1.08 5.36 25.0 4 0.69 6 97 polygon 1 ER 0.80

fanny 5.8 4.95 1.17 4.95 27.0 4 1.17 4 113 polygon 1 AN 0.52

fanny 7.33 4.69 1.56 4.69 107.0 5 0.6 9 107 polygon 1 EL 0.38

fanny 8.5 6.55 1.3 6.55 41.2 4 1.33 1 55.58 polygon 1 AF 0.54

fanny 12.59 11.88 1.06 11.88 117.5 3 1.23 12 104.46 spherical 1 EL 0.66

fanny 12.51 8.94 1.4 8.94 97.0 4 0.92 19 88.6 spherical 1 EL 0.93

fanny 4.18 3.92 1.07 3.92 15.3 4 0.7 5 85 polygon 1 ER 0.70

223
fanny 7.07 5.54 1.28 5.54 30.2 4 1.1 6 105 ovoid 1 AN 0.56

fanny 10.96 9.91 1.11 9.91 85.3 4 1.35 5 91.89 spherical 1 AN 0.47

fanny 9.77 8.43 1.16 8.43 61.6 3 0.74 8 94.15 spherical 1 ER 0.42

fanny 12.8 11.27 1.14 11.27 117.1 4 1.09 10 106 spherical 1 EL 0.48

fanny 8.5 7.25 1.17 7.25 59.9 4 0.94 15 73 spherical 1 EL 0.79

fanny 10.85 9.12 1.19 9.12 80.3 3 0.69 8 84 spherical 1 ER 0.33

fanny 11.7 9.09 1.29 9.09 90.3 4 0.94 9 96 polygon 1 EL 0.61

fanny 5.41 4.17 1.3 4.17 16.7 4 1.13 5 102 polygon 1 AN 0.44

fanny 8.49 6.85 1.24 6.85 49.0 3 0.88 8 88.08 polygon 1 EL 0.44

fanny 8.52 8.5 1 8.5 68.6 2 0.83 17 94.22 spherical 1 EL 0.75

fanny 8.98 8.9 1.01 8.9 63.3 2 0.92 8 72.47 spherical 1 ER 0.56

fanny 9.1 8.92 1.02 8.92 61.7 3 0.84 13 119 spherical 1 EL 0.51

fanny 12.92 10.41 1.24 10.41 98.1 3 1.02 21 103.14 ovoid 1 EL 0.99

fanny 8.32 7.2 1.16 7.2 49.2 3 0.83 5 97.88 polygon 2 ER 0.43

fanny 9.34 8.91 1.05 8.91 59.5 4 1.02 8 100 spherical 2 ER 0.40

fanny 13.92 13.83 1.01 13.83 137.4 4 1.14 19 112.56 ovoid 1 EL 0.80

fanny 13.93 13.11 1.06 13.11 144.2 2 1.19 17 56 spherical 1 EL 0.50

fanny 7.75 6.87 1.13 6.87 39.3 3 0.75 16 91.79 spherical 1 EL 0.55

fanny 10.9 8.08 1.35 8.08 67.4 3 0.88 8 101 spherical 1 ER 0.40

fanny 9.69 8.32 1.16 8.32 58.8 3 0.82 9 92 prolate 1 ER 0.50

fanny 14.74 13.03 1.13 13.03 85.6 4 0.92 6 85 spherical 1 EL 0.43

fanny 8.96 8.22 1.09 8.22 67.9 4 0.52 8 137 polygon 1 AF 0.39

fanny 17.98 15.2 1.18 15.2 234.6 3 1 14 97.16 spherical 1 EL 0.77

fanny 10.07 9.01 1.12 9.01 72.9 3 0.72 5 88.97 spherical 1 ER 0.41

fanny 13.4 9.02 1.49 9.02 96.4 4 0.66 8 116 ovoid 1 EL 0.41

fanny 9.69 8.74 1.11 8.74 72.8 3 0.78 13 111 spherical 1 EL 0.63

fanny 10.72 10.53 1.02 10.53 98.2 3 1.16 11 99 spherical 1 EL 0.43

fanny 5.4 4.72 1.14 4.72 19.6 3 0.6 6 83 spherical 1 ER 0.46

fanny 13.12 12.84 1.02 12.84 142.0 3 1.16 11 89.24 spherical 1 EL 0.58

fanny 4.95 4.27 1.16 4.27 17.3 3 0.7 4 17.3 spherical 1 AN 0.71

fanny 5.32 4.6 1.16 4.6 72.3 3 0.5 6 101 polygon 1 ER 0.39

fanny 6.38 3.33 1.92 3.33 17.9 3 0.42 7 102 ovoid 1 EL 0.35

fanny 10.4 9.45 1.1 9.45 73.0 4 0.72 13 92 ovoid 1 EL 0.96

fanny 10.89 9.04 1.2 9.04 77.2 3 0.6 11 81 ovoid 1 EL 0.92

fanny 4.67 3.76 1.24 3.76 11.8 3 0.47 8 96 spherical 1 ER 0.54

fanny 9.93 9.01 1.1 9.01 71.6 3 0.78 14 94 ovoid 1 EL 0.99

fanny 12.1 10.49 1.15 10.49 103.3 4 0.88 17 90.25 prolate 1 EL 0.99

fanny 9.04 8.85 1.02 8.85 59.4 2 0.84 8 97 spherical 1 ER 0.53

fanny 6.15 4.53 1.36 4.53 19.3 3 0.66 3 82 polygon 1 ER 0.39

fanny 10.21 9.21 1.11 9.21 76.4 2 0.74 18 110 spherical 1 EL 0.84

fanny 5.98 5.08 1.18 5.08 27.9 4 1 2 74 polygon 1 AN 0.44

fanny 11.06 10.44 1.06 10.44 90.5 3 0.82 10 117 spherical 1 ER 0.46

fanny 11.17 8.4 1.33 8.4 75.8 3 0.8 17 66 ovoid 1 EL 0.64

fanny 15.56 14.57 1.07 14.57 170.4 2 1.25 15 81.61 spherical 1 EL 0.85

224
fanny 7.62 6.31 1.21 6.31 40.4 3 0.72 12 89.9 prolate 1 EL 0.92

fanny 6.8 6.22 1.09 6.22 34.3 4 0.88 10 95.29 spherical 1 ER 0.82

fanny 8.77 7.89 1.11 7.89 55.2 3 0.87 9 99 spherical 1 ER 0.46

fanny 3.6 3.38 1.07 3.38 9.2 3 0.84 6 100 polygon 1 ER 0.86

fanny 5.22 4.92 1.06 4.92 19.5 3 0.7 5 93 spherical 1 ER 0.83

fanny 2.93 2.33 1.26 2.33 5.3 3 0.55 3 92 polygon 1 AN 0.45

fanny 4.31 3.89 1.11 3.89 12.0 3 0.55 3 99 spherical 1 AN 0.68

fanny 15 12.55 1.2 12.55 128.2 3 1.01 19 87 ovoid 1 EL 0.78

fanny 8.47 6.86 1.23 6.86 40.7 3 0.72 8 118 prolate 1 ER 0.58

fanny 6.88 4.59 1.5 4.59 33.4 4 0.97 6 105.71 polygon 1 ER 0.47

fanny 15.07 10.98 1.37 10.98 129.3 3 0.94 12 104.33 ovoid 1 EL 0.65

fanny 7.01 5.77 1.21 5.77 31.3 4 0.78 4 93 polygon 1 ER 0.48

fanny 15.18 14.77 1.03 14.77 186.3 3 1.14 7 97 spherical 1 EL 0.55

fanny 6.46 5.43 1.19 5.43 30.0 3 0.88 6 100 spherical 1 ER 0.68

fanny 9.06 8.09 1.12 8.09 61.4 3 0.88 8 82.57 spherical 1 ER 0.36

fanny 8.71 8.4 1.04 8.4 52.1 4 0.94 7 52.14 polygon 1 AF 0.61

fanny 6.25 4.52 1.38 4.52 22.7 3 0.8 9 94 ovoid 1 ER 0.86

fanny 10.71 6.96 1.54 6.96 66.6 3 1.1 12 95 prolate 1 EL 0.97

fanny 8.19 7.79 1.05 7.79 48.8 4 0.8 6 96 polygon 1 ER 0.50

fanny 8.48 8.41 1.01 8.41 57.6 3 0.93 7 86.42 spherical 1 ER 0.39

fanny 11.04 10.44 1.06 10.44 92.9 3 0.84 6 92.86 spherical 1 ER 0.44

fanny 18.84 16.9 1.11 16.9 261.6 4 1.05 7 124 spherical 1 EL 0.47

fanny 13.57 12.85 1.06 12.85 121.1 3 1 10 93.74 spherical 1 EL 0.56

fanny 13.26 12.89 1.03 12.89 149.9 3 1.6 12 92 polygon 1 EL 0.75

fanny 17.8 14.9 1.19 14.9 223.3 3 1.33 14 95 ovoid 1 EL 0.76

fanny 14.64 14.54 1.01 14.54 160.8 3 0.8 16 77 spherical 1 EL 0.83

fanny 6.1 6.05 1.01 6.05 31.4 4 0.65 6 114 spherical 1 ER 0.56

fanny 9.56 8.63 1.11 8.63 72.9 4 0.78 10 85.67 polygon 1 EL 0.50

fanny 8.5 8.09 1.05 8.09 52.3 4 1.1 7 101 polygon 1 EL 0.43

fanny 12.77 12.25 1.04 12.25 136.1 3 0.93 20 100 spherical 1 EL 0.86

fanny 12.69 12.09 1.05 12.09 120.8 3 0.94 12 98 spherical 1 EL 0.65

fanny 9.85 9.52 1.03 9.52 78.8 3 0.85 12 91 spherical 1 EL 0.46

fanny 10.32 9.45 1.09 9.45 67.7 5 0.75 11 112 polygon 1 EL 0.92

fanny 13.47 11.01 1.22 11.01 115.1 3 0.83 13 100 ovoid 1 EL 0.85

fanny 16.29 14.48 1.13 14.48 191.3 3 1 11 95 ovoid 1 EL 0.67

fanny 11.11 9.11 1.22 9.11 86.5 3 0.8 11 90 ovoid 1 EL 0.94

fanny 6.86 4.24 1.62 4.24 25.8 4 0.9 6 81 ovoid 1 AN 0.36

fanny 12.91 11.99 1.08 11.99 133.4 3 1.1 9 109 spherical 1 EL 0.44

fanny 5.4 4.2 1.29 4.2 16.2 4 0.72 5 76.85 spherical 1 ER 0.39

fanny 7.38 4.92 1.5 4.92 31.2 4 1 5 79 polygon 1 AN 0.45

fanny 4.06 3.98 1.02 3.98 15.0 4 0.72 5 65 spherical 1 AN 0.59

fanny 7.61 6.91 1.1 6.91 42.4 4 0.93 7 75 polygon 1 ER 0.52

fanny 6.45 5.25 1.23 5.25 31.2 4 0.92 9 108 ovoid 1 ER 0.72

fanny 10.3 9.39 1.1 9.39 79.1 3 0.92 12 93 ovoid 1 EL 0.95

225
fanny 5.02 4.41 1.14 4.41 17.7 3 1.05 4 71 spherical 1 AN 0.83

fanny 13.64 10.14 1.35 10.14 99.9 2 0.93 12 89 spherical 1 EL 0.59

fanny 9.36 8.61 1.09 8.61 63.7 3 0.51 9 120 ovoid 1 AF 0.38

fanny 5.61 4.54 1.24 4.54 16.2 4 0.8 6 102 polygon 1 ER 0.89

fanny 17.26 12.84 1.34 12.84 178.8 5 1.43 14 102 polygon 1 EL 0.74

fanny 17.55 13.1 1.34 13.1 190.7 3 1.3 12 100 prolate 1 EL 0.68

fanny 8.89 7.48 1.19 7.48 49.1 2 0.7 9 108 prolate 1 ER 0.54

fanny 3.38 3.28 1.03 3.28 11.7 3 0.46 6 75 polygon 1 ER 0.40

fanny 9.68 8.18 1.18 8.18 61.7 3 0.88 8 77 spherical 1 ER 0.34

fanny 13.95 12.1 1.15 12.1 126.8 3 0.87 17 95 ovoid 1 EL 0.77

fanny 11.27 10.36 1.09 10.36 93.0 2 0.87 9 112 spherical 1 ER 0.53

fanny 8.25 8.1 1.02 8.1 52.1 3 0.51 14 109 spherical 1 EL 0.71

fanny 5.04 4.3 1.17 4.3 15.7 3 0.62 4 113 spherical 1 ER 0.56

fanny 9.41 8.16 1.15 8.16 79.4 3 0.94 15 115 ovoid 1 EL 0.98

fanny 9.12 8.4 1.09 8.4 63.8 3 0.7 11 120 spherical 1 EL 0.53

fanny 9.22 7.47 1.23 7.47 59.0 4 1 8 92 spherical 1 AN 0.36

fanny 6.93 4.78 1.45 4.78 27.6 4 0.66 7 96 prolate 1 ER 0.51

fanny 9.22 7.88 1.17 7.88 57.8 4 0.58 9 103 ovoid 1 EL 0.41

fanny 15.86 15.72 1.01 15.72 215.0 4 1.17 24 90 ovoid 1 EL 0.82

fanny 8.55 6.95 1.23 6.95 52.7 4 0.8 10 83 prolate 1 ER 0.45

fanny 12.31 8.13 1.51 8.13 76.6 4 0.9 11 94 ovoid 1 EL 0.92

Brutus 5.92 5.78 1.02 5.78 33.0 3 0.66 6 91 polygon 1 ER 0.69

Brutus 5.08 4.81 1.06 4.81 19.2 3 0.69 9 90 spherical 1 ER 0.81

Brutus 6.46 5.36 1.21 5.36 31.1 4 0.75 10 85 polygon 1 ER 0.57

Brutus 6.86 5.23 1.31 5.23 26.0 4 0.72 5 118 ovoid 1 ER 0.68

Brutus 13.01 11.28 1.15 11.28 130.1 3 1.17 15 93 prolate 1 EL 0.99

Brutus 6.47 4.71 1.37 4.71 25.5 3 1.13 9 107 spherical 1 AN 0.58

Brutus 3.58 3.28 1.09 3.28 11.0 4 0.51 5 103 polygon 1 ER 0.48

Brutus 11.9 11.67 1.02 11.67 11.5 3 1.17 16 85 spherical 1 EL 0.81

Brutus 14.02 10.79 1.3 10.79 123.0 4 1.03 12 123 prolate 1 EL 0.65

Brutus 7.17 6.55 1.09 6.55 36.3 4 0.74 8 105 polygon 1 ER 0.79

Brutus 5.95 4.52 1.32 4.52 21.1 4 0.93 5 100 polygon 1 ER 0.74

Brutus 10.67 9.88 1.08 9.88 96.1 3 0.94 14 94 prolate 1 EL 0.98

Brutus 6.69 6.15 1.09 6.15 32.7 2 0.84 13 70 spherical 1 ER 0.51

Brutus 5.97 5.13 1.16 5.13 22.2 3 0.78 12 70 spherical 1 ER 0.54

Brutus 18.45 16.46 1.12 16.46 240.0 4 1.16 15 92 ovoid 1 EL 0.75

Brutus 11.22 9.11 1.23 9.11 81.2 4 0.75 15 111 prolate 1 EL 0.97

Brutus 24.76 17.74 1.4 17.74 341.7 3 1.7 19 102 ovoid 1 EL 0.78

Brutus 13.63 13.38 1.02 13.38 153.4 3 0.93 16 105 spherical 1 EL 0.84

Brutus 17.54 11.72 1.5 11.72 180.6 5 1.1 11 111 ovoid 1 EL 0.62

Brutus 13.98 11.82 1.18 11.82 140.7 3 0.94 10 90 spherical 1 EL 0.50

Brutus 18.93 13.04 1.45 13.04 205.1 3 1 19 86 prolate 1 EL 0.80

Brutus 7.18 6.16 1.17 6.16 40.0 3 0.8 10 89 spherical 1 ER 0.80

Brutus 9.45 6.78 1.39 6.78 48.8 4 0.8 11 88 prolate 1 EL 0.96

226
Brutus 18.62 14.62 1.27 14.62 208.0 3 1.05 14 105 ovoid 1 EL 0.75

Brutus 13.64 13.12 1.04 13.12 133.4 2 1.03 13 92 spherical 1 EL 0.82

Brutus 4.75 4.06 1.17 4.06 16.5 4 0.6 6 95 polygon 1 ER 0.71

Brutus 10.54 9.48 1.11 9.48 85.6 3 0.97 7 101 spherical 1 AN 0.35

Brutus 6.13 4.94 1.24 4.94 28.3 3 0.65 10 88 polygon 1 ER 0.67

Brutus 13.43 11.47 1.17 11.47 120.4 3 0.92 13 110 spherical 1 EL 0.75

Brutus 18.23 15.67 1.16 15.67 230.7 4 1.4 14 105 spherical 1 EL 0.76

Brutus 6.32 4.66 1.36 4.66 21.3 3 0.65 7 100 ovoid 1 ER 0.85

Brutus 13.26 9.86 1.34 9.86 113.3 4 1.14 10 113 prolate 1 EL 0.50

Brutus 12.95 9.75 1.33 9.75 86.2 3 1.17 10 91 spherical 1 EL 0.42

Brutus 7.85 7.81 1.01 7.81 42.3 5 0.66 8 113 ovoid 1 ER 0.55

Brutus 9.5 7.12 1.33 7.12 51.0 4 0.72 8 100 prolate 1 ER 0.45

Brutus 21 18 1.17 18 309.0 3 1.23 18 120 spherical 1 EL 0.82

Brutus 5.79 5.36 1.08 5.36 23.7 4 0.78 7 90 polygon 1 ER 0.84

Brutus 19.59 11.59 1.69 11.59 119.8 3 0.84 19 90 spherical 1 EL 0.70

Brutus 19 18.03 1.05 18.03 283.0 3 1.09 13 94 spherical 1 EL 0.81

Brutus 13.5 8.52 1.58 8.52 90.6 3 0.94 9 77.25 ovoid 1 EL 0.42

Brutus 5.94 5.43 1.09 5.43 23.3 3 0.72 6 86 polygon 1 ER 0.84

Brutus 6.35 3.38 1.88 3.38 16.8 4 0.72 7 89 quadrangular 1 ER 0.45

Brutus 5.57 5.37 1.04 5.37 26.6 3 0.78 7 86 polygon 1 ER 0.77

Brutus 11.9 10.66 1.12 10.66 105.0 4 0.9 13 100 ovoid 1 EL 0.97

Brutus 5.96 5.41 1.1 5.41 28.8 3 0.66 12 102 spherical 1 ER 0.60

Brutus 17.82 13.64 1.31 13.64 193.3 3 1.2 16 93 ovoid 1 EL 0.79

Brutus 9.43 7.62 1.24 7.62 56.0 3 0.77 11 97 prolate 1 EL 0.95

Brutus 7.75 5.69 1.36 5.69 33.2 4 0.92 9 99 polygon 1 ER 0.55

Brutus 5.04 4.45 1.13 4.45 18.9 4 0.6 7 78 polygon 1 ER 0.53

Brutus 9.45 7.25 1.3 7.25 47.0 3 0.78 12 88 ovoid 1 EL 0.98

Brutus 10.44 6.43 1.62 6.43 45.8 5 0.78 6 100 ovoid 1 ER 0.35

Brutus 6.7 4.8 1.4 4.8 25.2 3 0.65 8 101 prolate 1 ER 0.62

Brutus 12.08 7.07 1.71 7.07 68.8 3 1.2 7 57 prolate concave-convex 1 SA 0.39

Brutus 8.45 6.91 1.22 6.91 43.1 4 0.5 10 93.6 triangular 1 AF 0.36

Brutus 13.57 9.16 1.48 9.16 101.6 4 1.03 9 111 elongate 1 EL 0.43

Brutus 25.24 11.43 2.21 11.43 139.4 3 1.37 8 119 prolate 1 SA 0.46

Brutus 6.16 4.05 1.52 4.05 20.7 3 1 7 73.26 prolate 1 AN 0.58

Brutus 6.5 5.32 1.22 5.32 28.0 3 0.69 8 94 spherical 1 ER 0.84

Brutus 5.87 5.69 1.03 5.69 27.2 3 0.65 5 96 spherical 1 ER 0.66

Brutus 4.56 3.19 1.43 3.19 14.1 4 0.61 6 78 polygon 1 AN 0.42

Brutus 6.78 5.45 1.24 5.45 29.9 5 1.03 5 82 polygon 1 AN 0.45

Brutus 5.39 3.67 1.47 3.67 17.2 4 0.94 6 89 polygon 1 ER 0.48

Brutus 6.19 5.18 1.19 5.18 27.7 4 0.8 7 78 polygon 1 ER 0.62

Brutus 7.06 5.85 1.21 5.85 35.6 3 0.8 12 93 prolate 1 EL 0.93

Brutus 5.43 4.31 1.26 4.31 18.1 4 0.85 9 94 prolate 1 ER 0.86

Brutus 9.11 7.83 1.16 7.83 57.1 2 0.93 13 94 spherical 1 EL 0.58

Brutus 10.28 9.56 1.08 9.56 74.7 4 1.13 17 113 ovoid 1 EL 0.97

227
Brutus 10.78 9.75 1.11 9.75 90.0 3 0.92 11 97 spherical 1 EL 0.53

Brutus 10.86 9.85 1.1 9.85 84.5 4 0.84 14 108 ovoid 1 EL 0.97

Brutus 15.01 14.76 1.02 14.76 185.8 3 0.97 15 107 spherical 1 EL 0.81

Brutus 8.3 6.09 1.36 6.09 90.8 4 1.01 9 90 polygon 1 EL 0.39

Brutus 14.17 14.01 1.01 14.01 150.5 4 1.01 13 85.85 spherical 1 EL 0.76

Brutus 16.81 14.98 1.12 14.98 194.0 3 0.88 14 101.75 spherical 1 EL 0.77

Brutus 16.35 15.4 1.06 15.4 202.0 2 1.4 13 110 spherical 1 EL 0.85

Brutus 8.15 7.3 1.12 7.3 43.0 3 0.66 10 99 spherical 1 ER 0.65

Brutus 5.82 5.69 1.02 5.69 31.4 3 0.64 7 97 polygon 1 ER 0.53

Brutus 14.8 14.2 1.04 14.2 165.7 3 1.19 10 115.7 spherical 1 EL 0.54

Brutus 6.33 5.71 1.11 5.71 25.7 3 0.6 9 110 spherical 1 ER 0.57

Brutus 5.68 4.95 1.15 4.95 26.5 4 0.74 6 92.7 polygon 1 ER 0.90

Brutus 6.95 5.95 1.17 5.95 34.6 3 0.66 10 100 polygon 1 ER 0.79

Brutus 8.66 8.32 1.04 8.32 56.0 3 0.83 10 102 polygon 1 ER 0.48

Brutus 17.72 13.52 1.31 13.52 159.0 3 1.02 15 120 ovoid 1 EL 0.74

Brutus 8.37 6.22 1.35 6.22 34.2 4 0.55 8 120 ovoid 1 AF 0.38

Brutus 10.27 6.71 1.53 6.71 60.1 4 0.94 9 97 prolate concave-convex 1 EL 0.42

Brutus 11.95 8.36 1.43 8.36 82.0 3 1.14 12 85 ovoid 1 EL 0.92

Brutus 11.43 10.58 1.08 10.58 103.7 3 1.03 15 87 spherical 1 EL 0.95

Brutus 10.14 9.01 1.13 9.01 75.0 5 1.02 9 96.6 polygon 1 EL 0.44

Brutus 9.02 7.79 1.16 7.79 58.5 3 0.83 11 100 spherical 1 EL 0.50

Brutus 14.24 11.86 1.2 11.86 123.8 3 0.92 17 81 prolate 1 EL 0.80

Brutus 11.39 10.21 1.12 10.21 91.5 3 0.82 19 86 ovoid 1 EL 0.97

Brutus 14.57 14.01 1.04 14.01 169.3 3 1.14 15 116 spherical 1 EL 0.83

Brutus 12.44 9.75 1.28 9.75 99.7 3 1.13 10 97 ovoid 1 EL 0.49

Brutus 5.86 4.52 1.3 4.52 26.3 4 0.6 6 109.4 polygon 1 ER 0.59

hector 6.32 6.04 1.05 6.04 30.5 4 0.83 7 80 polygon 1 ER 0.60

hector 7.17 4.82 1.49 4.82 30.3 4 0.7 9 101 ovoid 1 ER 0.46

hector 6.59 4.76 1.38 4.76 25.8 4 0.8 7 115 ovoid 1 ER 0.79

hector 5.95 3.79 1.57 3.79 18.5 3 0.72 5 97 prolate 1 ER 0.65

hector 11.66 10.63 1.1 10.63 110.0 4 0.97 6 95 spherical 1 EL 0.44

hector 19.32 15.84 1.22 15.84 281.6 1 2.4 15 77.5 prolate 1 EL 0.78

hector 8.24 8.12 1.01 8.12 53.4 2 0.7 12 90 spherical 1 EL 0.41

hector 5.92 5.91 1 5.91 27.9 3 0.72 9 98 spherical 1 ER 0.49

hector 7.65 4.95 1.55 4.95 32.6 4 0.72 4 70 prolate 1 AN 0.42

hector 6.03 3.98 1.52 3.98 27.3 4 0.94 4 77 polygon 1 AN 0.47

hector 4.02 3.5 1.15 3.5 11.4 4 0.83 4 80 polygon 1 ER 0.52

hector 20.32 14.28 1.42 14.28 228.7 3 0.9 30 78 prolate 1 EL 0.81

hector 7.97 7.03 1.13 7.03 43.6 3 1 7 87 ovoid 1 AN 0.52

hector 5.14 4.15 1.24 4.15 16.6 4 0.52 4 100 ovoid 1 AN 0.85

hector 12.98 10.1 1.29 10.1 119.1 4 1.3 8 80.41 ovoid 1 EL 0.47

hector 16.7 13.62 1.23 13.62 198.1 3 1.14 17 71 ovoid 1 EL 0.74

hector 12.6 11.75 1.07 11.75 114.4 3 1 4 120 spherical 1 AN 0.31

hector 12 9.68 1.24 9.68 107.0 4 1 5 100 spherical 1 EL 0.37

228
hector 19.7 16.43 1.2 16.43 254.7 3 1.44 9 102 ovoid 1 EL 0.47

hector 8.15 7.37 1.11 7.37 53.5 4 1.13 4 84 polygon 1 AN 0.55

hector 12.76 10.63 1.2 10.63 102.6 4 0.87 9 100 spherical 1 EL 0.53

kendo 6.31 4.69 1.35 4.69 23.0 5 0.72 5 88 polygon 1 ER 0.82

kendo 6.45 5.43 1.19 5.43 29.5 3 0.94 7 82.55 spherical 1 AF 0.36

kendo 7.72 5.05 1.53 5.05 42.5 2 1.17 11 96.81 prolate 1 EL 0.90

kendo 15.35 13.95 1.1 13.95 191.8 4 1.49 5 84.4 polygon 1 SA 0.45

kendo 4.37 4.04 1.08 4.04 16.4 5 0.5 5 88 polygon 1 AN 0.48

kendo 7.59 6.16 1.23 6.16 34.0 3 0.72 13 83.68 spherical 1 ER 0.55

kendo 3.88 3.2 1.21 3.2 9.4 3 0.81 6 92.55 polygon 1 ER 0.91

kendo 16.78 14.13 1.19 14.13 175.7 3 1.03 12 107 ovoid 1 EL 0.68

kendo 4.2 4.04 1.04 4.04 23.6 4 0.98 6 0.81 polygon 1 AN 0.50

kendo 11.01 10.42 1.06 10.42 98.0 3 0.9 14 94 spherical 1 EL 0.85

kendo 6.67 6.31 1.06 6.31 39.4 4 0.74 9 94 polygon 1 ER 0.81

kendo 5.7 5.02 1.14 5.02 26.0 3 0.9 7 98 spherical 1 ER 0.73

kendo 14.84 13.38 1.11 13.38 161.6 3 1.03 8 111 spherical 1 EL 0.48

kendo 7.1 6.77 1.05 6.77 41.4 4 1.17 8 97 polygon 1 ER 0.39

kendo 5.18 4.68 1.11 4.68 23.7 3 0.98 4 102 polygon 1 AN 0.51

kendo 7.85 7.1 1.11 7.1 42.4 3 0.81 8 91 polygon 1 ER 0.72

Ondine 4.85 4.5 1.08 4.5 16.4 2 0.55 7 110 spherical 1 ER 0.85

Ondine 4.43 3.19 1.39 3.19 12.0 4 0.46 7 90 polygon 1 ER 0.41

Ondine 8.23 7.7 1.07 7.7 55.2 3 0.66 10 90 spherical 1 ER 0.49

Ondine 7.17 6.56 1.09 6.56 39.1 3 1.05 15 70 spherical 1 EL 0.56

Ondine 5.18 4.24 1.22 4.24 18.2 2 0.62 5 102 prolate 1 ER 0.83

Ondine 8.92 7.23 1.23 7.23 52.7 2 0.8 14 85 ovoid 1 EL 0.99

Ondine 6.96 3.69 1.89 3.69 22.2 3 0.72 7 99 polygon 1 EL 0.47

Ondine 6.34 4.67 1.36 4.67 29.5 3 0.66 6 120 prolate concave-convex 1 ER 0.75

Ondine 4.57 3.1 1.47 3.1 12.0 5 0.72 7 106 polygon 1 ER 0.60

Lefkas 5.53 4.27 1.3 4.27 21.7 4 0.78 6 118 spherical 1 ER 0.78

Lefkas 4.71 4.3 1.1 4.3 15.5 3 0.51 9 87 polygon 1 ER 0.56

Lefkas 7.56 6.06 1.25 6.06 38.2 4 0.81 11 105 spherical 1 ER 0.55

Lefkas 6.19 4.64 1.33 4.64 26.9 4 0.75 9 94 polygon 1 ER 0.91

Lefkas 6.41 5.34 1.2 5.34 24.1 4 0.58 9 75 ovoid 1 AF 0.38

Lefkas 4.64 2.79 1.66 2.79 9.8 5 0.72 6 93 polygon 1 ER 0.58

Lefkas 6.46 5.85 1.1 5.85 24.8 3 0.62 6 99 spherical 1 ER 0.68

Lefkas 6.5 5.18 1.25 5.18 25.0 3 0.66 14 96 spherical 1 EL 0.58

Lefkas 5.76 4.95 1.16 4.95 24.9 4 0.88 8 71 polygon 1 ER 0.44

Lefkas 5.84 4.81 1.21 4.81 22.3 3 0.94 8 100 spherical 1 ER 0.80

Lefkas 14.44 14.14 1.02 14.14 173.3 2 0.97 11 116 spherical 1 EL 0.64

Agathe 9.99 8.36 1.19 8.36 67.6 3 0.66 11 101 ovoid 1 EL 0.95

Agathe 6.79 5.14 1.32 5.14 29.0 3 0.51 7 93 prolate 1 AF 0.54

Agathe 7.3 6.09 1.2 6.09 34.6 5 0.92 5 92 polygon 1 ER 0.66

Agathe 5.43 5.07 1.07 5.07 23.3 4 0.58 10 93 polygon 1 ER 0.52

Agathe 7.2 5.67 1.27 5.67 27.7 4 0.92 7 95 ovoid 1 ER 0.61

229
Agathe 14.45 9.62 1.5 9.62 108.9 3 1.09 9 109 ovoid 1 EL 0.42

Agathe 8.33 7.35 1.13 7.35 44.6 4 0.8 10 96 polygon 1 ER 0.50

Agathe 12.27 11.67 1.05 11.67 127.7 4 1.09 24 88 ovoid 1 EL 0.96

Agathe 4.6 3.94 1.17 3.94 16.0 3 0.72 8 85 spherical 1 ER 0.73

Agathe 7.36 5.13 1.43 5.13 32.2 4 0.83 8 104 prolate 1 ER 0.53

Agathe 13.39 11.78 1.14 11.78 129.8 2 0.92 9 125 prolate 1 EL 0.51

Agathe 5.84 5.02 1.16 5.02 29.9 3 0.58 8 110 polygon 1 ER 0.49

Agathe 3.94 3.86 1.02 3.86 14.5 4 0.58 9 79 polygon 1 ER 0.49

Agathe 4.66 4.64 1 4.64 16.8 3 0.8 9 92 spherical 1 ER 0.56

Agathe 7.58 6.46 1.17 6.46 43.2 3 0.65 6 110 spherical 1 ER 0.58

Agathe 18.04 16.72 1.08 16.72 241.0 2 1.23 16 88 ovoid 1 EL 0.86

Agathe 9.85 6.25 1.58 6.25 45.7 4 0.94 9 98 prolate 1 EL 0.37

Agathe 5.16 5.02 1.03 5.02 23.6 3 0.51 8 91 polygon 1 ER 0.52

Agathe 8.71 6.36 1.37 6.36 49.0 3 0.72 9 117 ovoid 1 ER 0.48

Agathe 9.88 8.61 1.15 8.61 76.8 3 1.09 13 81.75 spherical 1 EL 0.68

Agathe 3.79 3.69 1.03 3.69 12.9 4 0.72 5 114 polygon 1 ER 0.75

Agathe 6.37 4.32 1.47 4.32 22.2 4 0.65 10 93 ovoid 1 ER 0.44

Agathe 4.64 3.13 1.48 3.13 14.4 4 0.46 5 106 polygon 1 AN 0.45

Agathe 4.03 3.87 1.04 3.87 14.5 4 0.78 6 95 polygon 1 ER 0.84

Agathe 10.78 9.49 1.14 9.49 78.7 2 0.72 9 105 spherical 1 ER 0.50

Agathe 4.82 3.49 1.38 3.49 16.2 4 0.8 5 87 polygon 1 ER 0.66

Agathe 12.37 7.33 1.69 7.33 76.5 2 1.85 11 76 prolate 1 EL 0.73

Agathe 4.24 3.62 1.17 3.62 11.9 4 0.62 6 73 polygon 1 ER 0.56

Agathe 2.81 2.48 1.13 2.48 4.9 3 0.5 4 86 polygon 1 AN 0.39

Agathe 4.61 3.28 1.41 3.28 13.1 3 0.62 7 92 polygon 1 ER 0.58

Agathe 6.49 5.87 1.11 5.87 29.8 2 0.92 8 92.35 spherical 1 ER 0.86

Agathe 5.24 5.23 1 5.23 20.8 4 0.69 8 100 polygon 1 ER 0.49

Agathe 3.71 3.67 1.01 3.67 10.4 3 0.8 4 82 spherical 1 ER 0.58

Agathe 14.06 8.63 1.63 8.63 99.9 2 1.03 11 70 ovoid 1 EL 0.51

Agathe 13.98 11.8 1.18 11.8 121.5 4 1.3 10 100 prolate 1 EL 0.47

Agathe 3.08 2.67 1.15 2.67 6.6 3 0.55 9 82 spherical 1 ER 0.43

Agathe 6.76 5.23 1.29 5.23 30.6 3 0.84 6 108 prolate 1 ER 0.79

Agathe 9.87 5.92 1.67 5.92 49.1 4 0.69 10 100 ovoid 1 EL 0.51

Agathe 9.91 8.01 1.24 8.01 61.0 3 0.62 8 86 ovoid 1 EL 0.40

Agathe 5.74 4.12 1.39 4.12 21.4 4 0.87 7 93.61 polygon 1 ER 0.67

Agathe 8.35 5.53 1.51 5.53 39.5 4 0.83 8 107 ovoid 1 ER 0.39

Agathe 6.54 5 1.31 5 75.5 3 0.84 7 75.5 spherical 1 AF 0.43

Agathe 10.96 7.11 1.54 7.11 55.1 3 0.92 9 66.1 prolate 1 AF 0.42

Agathe 9.85 7.58 1.3 7.58 61.3 3 0.94 7 91.53 spherical 1 ER 0.38

Agathe 5.59 5.1 1.1 5.1 24.2 4 0.72 9 86 spherical 1 ER 0.76

Agathe 5.24 4.31 1.22 4.31 17.9 3 0.6 7 87 spherical 1 ER 0.67

Agathe 4.61 4.3 1.07 4.3 17.6 4 0.72 5 103.17 polygon 1 ER 0.77

Agathe 8.92 8.5 1.05 8.5 55.3 2 0.84 9 97.72 spherical 1 ER 0.55

Agathe 10.01 5.95 1.68 5.95 48.2 5 1.02 8 90.1 ovoid 1 AN 0.41

230
Agathe 11.26 10.14 1.11 10.14 98.3 3 1.47 14 71.23 spherical 1 EL 0.88

Agathe 5.23 3.38 1.55 3.38 14.8 4 0.83 7 99.55 polygon 1 ER 0.62

Agathe 11.04 7 1.58 7 68.6 4 0.74 10 94.7 ovoid 1 EL 0.49

Clyde 8.54 7.26 1.18 7.26 47.3 3 0.74 11 87.78 prolate 1 EL 0.93

Clyde 7.1 5 1.42 5 31.1 2 0.51 12 113.61 prolate 1 EL 0.91

Clyde 3.65 3.23 1.13 3.23 9.4 3 0.51 5 91 polygon 1 ER 0.47

Clyde 4.77 4.17 1.14 4.17 16.9 3 0.55 8 93 polygon 1 ER 0.59

Clyde 4.85 3.4 1.43 3.4 14.8 3 0.74 6 93 polygon 1 ER 0.64

Clyde 7.79 6.96 1.12 6.96 47.5 4 0.69 10 88 ovoid 1 ER 0.52

Clyde 5.43 3.5 1.55 3.5 16.2 3 0.72 6 74 prolate 1 AN 0.35

Clyde 3.94 2.47 1.6 2.47 8.1 4 0.51 6 77 polygon 1 EL 0.38

Clyde 3.99 3.16 1.26 3.16 7.8 4 0.51 4 79 polygon 1 AN 0.58

Tina 8.42 5.14 1.64 5.14 43.1 4 0.82 11 96.75 ovoid 1 EL 0.93

Tina 3.69 2.46 1.5 2.46 7.6 3 0.6 6 91 polygon 1 ER 0.49

Tina 6.64 5.67 1.17 5.67 31.6 4 0.72 13 85.65 spherical 1 ER 0.55

Tina 8.7 7.79 1.12 7.79 45.2 4 0.72 8 45.22 polygon 1 AF 0.92

Tina 9.11 7.68 1.19 7.68 54.6 5 1.31 6 68.12 polygon 1 EL 0.46

Tina 8.85 4.92 1.8 4.92 37.9 4 0.87 5 98.19 prolate 1 EL 0.37

Tina 5.94 3.38 1.76 3.38 18.7 4 0.69 4 108.59 polygon 1 ER 0.35

Tina 4.05 2.66 1.52 2.66 11.2 4 0.61 8 100 prolate 1 ER 0.55

Tina 7.58 7.28 1.04 7.28 44.5 3 0.94 11 89.22 spherical 1 ER 0.61

Mkubwa 3.49 3.18 1.1 3.18 8.7 3 0.5 6 100 spherical 1 ER 0.57

Mkubwa 7.84 6.24 1.26 6.24 46.6 3 1.02 9 91.05 spherical 1 AN 0.45

Mkubwa 2.88 2.34 1.23 2.34 6.1 3 0.58 5 102 spherical 1 ER 0.53

Oreste 8.7 7.17 1.21 7.17 57.1 3 0.72 9 94.11 quadrangular 1 ER 0.47

Oreste 11.58 9.44 1.23 9.44 89.0 3 1.04 14 97.62 ovoid 1 EL 0.96

Oreste 5.97 5.34 1.12 5.34 24.9 4 0.61 8 101 polygon 1 ER 0.75

Oreste 4.04 3.01 1.34 3.01 16.0 4 0.82 6 75.75 polygon 1 ER 0.55

Oreste 7.2 6.57 1.1 6.57 35.4 2 0.6 6 105 spherical 1 ER 0.62

Oreste 9.4 9.32 1.01 9.32 73.7 2 0.85 7 97.79 spherical 1 ER 0.48

Oreste 14.64 12.05 1.21 12.05 133.2 2 0.83 13 114.75 prolate 1 EL 0.75

Oreste 13.37 10.73 1.25 10.73 115.5 3 0.9 12 128 prolate 1 EL 0.81

Oreste 8.11 6.5 1.25 6.5 42.4 4 0.6 7 95 prolate 1 ER 0.50

Oreste 12.89 11.2 1.15 11.2 118.4 3 0.52 13 87 prolate 1 EL 0.92

Oreste 12.42 7.79 1.59 7.79 74.2 4 0.94 12 102 ovoid 1 EL 0.91

Oreste 7.85 6.66 1.18 6.66 40.0 2 0.62 10 123 prolate 1 ER 0.67

Oreste 14.42 12.09 1.19 12.09 139.2 2 0.72 17 122 prolate 1 EL 0.79

Oreste 6.97 6.82 1.02 6.82 39.7 4 0.8 6 87 spherical 1 ER 0.70

Oreste 12.98 12.46 1.04 12.46 138.3 5 1.16 20 90 polygon 1 EL 0.89

Oreste 5.17 3.95 1.31 3.95 21.4 3 0 0 0 polygon 1 AF 0.71

Oreste 9.21 7.01 1.31 7.01 50.5 4 0.75 9 99 polygon 1 EL 0.48

Oreste 6.26 5.86 1.07 5.86 26.7 1 0.4 6 117 spherical 1 AF 0.46

Oreste 6.56 6.24 1.05 6.24 30.5 3 0.72 8 117 polygon 1 ER 0.70

Oreste 7.07 6.99 1.01 6.99 40.1 2 0.46 11 124 spherical 1 ER 0.63

231
Oreste 13.75 11.53 1.19 11.53 133.8 4 0.94 20 92.32 spherical 1 EL 0.79

Oreste 18.6 14.65 1.27 14.65 219.9 3 0.74 23 115 prolate 1 EL 0.78

Oreste 11.39 9.26 1.23 9.26 86.8 4 2.1 4 70 polygon 1 EL 0.51

Oreste 9.32 8.19 1.14 8.19 64.2 5 1.33 8 91 polygon 1 EL 0.43

Oreste 11.63 9.53 1.22 9.53 81.6 4 0.9 11 116 ovoid 1 EL 0.91

Oreste 9.91 7.01 1.41 7.01 64.1 4 0.91 10 89.35 quadrangular 1 EL 0.42

Oreste 5.06 4.03 1.26 4.03 16.1 4 0.62 7 81 polygon 1 ER 0.60

Oreste 18.34 14.28 1.28 14.28 165.9 5 0.72 8 97 polygon 1 EL 0.49

Oreste 9.32 7.38 1.26 7.38 60.0 4 0.72 15 89 polygon 2 EL 0.99

Oreste 10.52 9.54 1.1 9.54 81.6 3 0.92 13 91.63 prolate 2 EL 0.96

Oreste 10.51 9.49 1.11 9.49 87.4 3 1.1 9 112.25 ovoid 1 AN 0.35

Oreste 12.72 7.12 1.79 7.12 52.6 3 0 0 0 angular point 1 SA 0.98

Oreste 13.62 11.74 1.16 11.74 131.7 3 1.1 6 58 polygon 1 SA 0.65

Oreste 6.35 5.43 1.17 5.43 29.2 5 0.7 8 95 polygon 1 ER 0.81

Oreste 14.65 11.98 1.22 11.98 155.4 3 2.36 6 91 polygon 1 EL 0.48

Starch microremains from calculus. ER=Eremospatha, AF=Aframomum, AN=Laccosperma, GI=Gilbertiodendron, CO=Cola, NA=Napoleona,
TR=Treculia, CU=Coula, XY=Xylia, PI=Piper, PA=Panda, SG=Sacoglottis, CL=Calpocalyx.
Chimpanzee name

Length

Width

LW Ratio

Brea

Area

Shape

Facets

Striaelen

Striaeno

Type

Lam

Dist

certainty score
Genus with highest

Certainty score
castor 13.21 12.67 1.0 12.7 131.22 spherical 0 0 0 1 1 6.16 GI 0.31

castor 16.89 14.11 1.2 14.1 191.72 ovoid 0 0 0 1 2 10.78 CO 0.45

bijou 6.02 4.98 1.2 5.0 24.77 spherical 0 0 0 1 0 3.01 NA 0.30

bijou 12.38 10.67 1.2 10.7 113.42 spherical 0 0 0 1 0 7.3 GI 0.32

bijou 11.67 11.16 1.0 11.2 103.46 spherical 0 2.2 2 1 0 5.12 GI 0.28

bijou 11.14 7.47 1.5 7.5 69.21 ovoid 0 0 0 1 0 6.45 TR 0.51

bijou 8.64 6.53 1.3 6.5 59.32 ovoid 0 0 0 1 1 3.98 TR 0.66

bijou 5.63 4.92 1.1 4.9 23.65 spherical 1 0 0 1 1 2.78 CU 0.28

bijou 5.12 5.12 1.0 5.0 20.55 spherical 1 0 0 1 1 1.96 ER 0.33

bijou 9.76 9.11 1.1 9.1 81.09 spherical 1 1.27 2 1 0 4.88 GI 0.39

bijou 10.26 10.26 1.0 10.2 82.18 spherical 0 0 0 1 0 5.13 CU 0.35

fanny 10.03 8.08 1.2 8.1 56.48 ovoid 0 0 0 1 0 5.32 TR 0.50

fanny 5.01 4.9 1.0 4.9 19.72 polygon 7 0 0 3 0 1.95 PI 0.53

fanny 3.71 3.5 1.1 3.5 10.52 hemispherical 1 0 0 1 0 1.24 XY 0.31

fanny 5.25 5.25 1.0 5.0 26.12 spherical 0 0 0 1 0 1.85 ER 0.31

fanny 4.83 4.32 1.1 4.3 16.57 spherical 1 0 0 1 0 1.74 NA 0.24

fanny 11.38 11.18 1.0 11.2 108.01 oblate conovoid 2 1.59 1 2 0 5.69 GI 0.41

fanny 3.62 3.33 1.1 3.3 10.84 hemispherical 1 0 0 1 0 1.54 XY 0.32

fanny 12.18 11.48 1.1 11.5 115.41 spherical 0 0 0 1 0 5.47 GI 0.34

fanny 20.04 16.43 1.2 16.4 242.33 polygon 6 0 0 1 1 8.47 SA 0.63

fanny 8.82 8.69 1.0 8.7 56.17 oblate conovoid 3 0 0 2 0 2.77 GI 0.45

fanny 5.99 5.51 1.1 5.5 26.15 spherical 0 0 0 1 1 2.15 NA 0.25

232
fanny 15.52 11.89 1.3 11.9 132.17 ovoid 0 3.2 2 1 0 6.06 CO 0.40

fanny 9.35 8.4 1.1 8.4 62.36 oblate conovoid 3 0 0 1 0 4.675 GI 0.43

fanny 6.06 5.73 1.1 5.7 33.11 hemispherical 1 0 0 1 0 2.36 GI 0.27

fanny 7.7 7.08 1.1 7.1 43.67 spherical 0 0 0 1 0 2.1 ER 0.21

fanny 2.43 2.14 1.1 2.1 5.09 oblate conovoid 1 0 0 1 0 1.04 CL 0.33

fanny 10.14 9.14 1.1 9.1 81.11 spherical 0 4.5 3 1 0 4.71 GI 0.27

fanny 4.94 4.51 1.1 4.5 18.28 spherical 0 0 0 1 0 1.44 ER 0.26

fanny 8.43 6.48 1.3 6.5 37.77 prolate 0 0 0 1 0 3.8 TR 0.32

fanny 4.86 4.22 1.2 4.2 18.97 polygon 7 0 0 3 0 2.63 PI 0.55

fanny 6.02 4.78 1.3 4.8 20.54 polygon 7 0 0 3 0 3.01 PI 0.46

fanny 4.86 3.5 1.4 3.5 16.57 polygon 7 0 0 3 0 2 PI 0.72

fanny 4.36 3.25 1.3 3.3 18.98 polygon 7 0 0 3 0 1.75 PI 0.70

fanny 6.22 5.18 1.2 5.2 31.03 polygon 5 0 0 3 0 2.19 PI 0.40

fanny 6.36 3.48 1.8 3.5 27.6 polygon 8 0 0 3 0 2.87 PI 0.51

fanny 5.04 3.75 1.3 3.8 35.22 polygon 7 0 0 3 0 2.52 PI 0.54

fanny 7.6 6.78 1.1 6.8 36.75 polygon 8 0 0 3 0 3.58 SA 0.57

fanny 4.61 4.61 1.0 4.6 18.18 spherical 0 0 0 1 1 1.23 ER 0.47

fanny 9.3 9.3 1.0 8.4 64.03 spherical 0 0 0 1 0 4.31 CU 0.41

fanny 4.86 4.4 1.1 4.4 17.47 polygon 7 0 0 3 0 2.43 PI 0.60

fanny 5.22 4.61 1.1 4.6 19.72 polygon 5 0 0 3 0 2.31 PI 0.52

fanny 4.74 4.64 1.0 4.6 14.72 polygon 8 0 0 3 0 2.14 PI 0.58

fanny 4.32 4.14 1.0 4.1 18.99 polygon 6 0 0 3 0 1.85 PI 0.58

fanny 3.53 3.18 1.1 3.2 15.66 polygon 3 0 0 3 0 2.15 PI 0.53

fanny 6.51 4.12 1.6 4.1 18.44 polygon 5 0 0 3 0 3.255 PI 0.39

concaveconvex
fanny 5.2 3.69 1.4 3.7 17.72 polygon 7 0 0 3 0 2.6 PI 0.67

fanny 3.67 3.13 1.2 3.1 10.76 polygon 7 0 0 3 0 1.835 PI 0.74

fanny 4.67 4.46 1.0 4.5 20.97 polygon 7 0 0 3 0 2.335 PI 0.53

Leo 14.23 12.29 1.2 12.3 130.05 spherical 2 0 0 1 1 7.43 GI 0.46

Leo 15.02 14.2 1.1 14.2 178 spherical 0 2.03 1 1 0 7.99 GI 0.29

Leo 22.91 18.17 1.3 18.2 346 prolate 0 9.44 2 1 3 7.95 CO 0.73

Leo 19.42 13.85 1.4 13.9 218 ovoid 0 0 0 1 0 9.5 GI 0.39

Leo 6.73 6.12 1.1 6.1 38.18 spherical 0 0 0 1 0 2.56 NA 0.23

castor 5.5 5.08 1.1 5.1 24.45 polygon 8 0 0 3 0 2.19 PI 0.42

castor 5.5 5.08 1.1 5.1 24.45 polygon 8 0 0 3 0 2.19 PI 0.42

castor 5.5 5.08 1.1 5.1 24.45 polygon 8 0 0 3 0 2.19 PI 0.42

castor 5.5 5.08 1.1 5.1 24.45 polygon 8 0 0 3 0 2.19 PI 0.42

castor 5.5 5.08 1.1 5.1 24.45 polygon 8 0 0 3 0 2.19 PI 0.42

castor 5.5 5.08 1.1 5.1 24.45 polygon 8 0 0 3 0 2.19 PI 0.42

castor 5.5 5.08 1.1 5.1 24.45 polygon 8 0 0 3 0 2.19 PI 0.42

castor 5.5 5.08 1.1 5.1 24.45 polygon 8 0 0 3 0 2.19 PI 0.42

castor 5.5 5.08 1.1 5.1 24.45 polygon 8 0 0 3 0 2.19 PI 0.42

castor 5.5 5.08 1.1 5.1 24.45 polygon 8 0 0 3 0 2.19 PI 0.42

castor 5.5 5.08 1.1 5.1 24.45 polygon 8 0 0 3 0 2.19 PI 0.42

castor 5.5 5.08 1.1 5.1 24.45 polygon 8 0 0 3 0 2.19 PI 0.42

castor 5.5 5.08 1.1 5.1 24.45 polygon 8 0 0 3 0 2.19 PI 0.42

castor 5.5 5.08 1.1 5.1 24.45 polygon 8 0 0 3 0 2.19 PI 0.42

233
castor 5.5 5.08 1.1 5.1 24.45 polygon 8 0 0 3 0 2.19 PI 0.42

castor 5.5 5.08 1.1 5.1 24.45 polygon 8 0 0 3 0 2.19 PI 0.42

castor 5.5 5.08 1.1 5.1 24.45 polygon 8 0 0 3 0 2.19 PI 0.42

castor 5.5 5.08 1.1 5.1 24.45 polygon 8 0 0 3 0 2.19 PI 0.42

castor 5.5 5.08 1.1 5.1 24.45 polygon 8 0 0 3 0 2.19 PI 0.42

castor 5.5 5.08 1.1 5.1 24.45 polygon 8 0 0 3 0 2.19 PI 0.42

castor 6.41 4.36 1.5 4.4 24.45 polygon concave 8 0 0 3 0 2.19 PI 0.36
convex
castor 13.83 11.75 1.2 11.8 113.82 triangular 0 0 0 1 1 6 CO 0.45

Goma 24.38 19.47 1.3 19.5 342 ovoid 0 0 0 1 2 17.96 CO 0.57

Goma 5.32 4.92 1.1 4.9 25.84 spherical 0 0 0 1 0 2.61 NA 0.34

Goma 4.38 3.67 1.2 3.7 14.85 spherical 0 0 0 1 0 1.95 ER 0.29

Goma 29.37 20.49 1.4 20.5 432 ovoid 0 0 0 1 2 20 CO 0.58

Goma 12.42 11.61 1.1 11.6 107 spherical 0 3 2 1 1 4.82 GI 0.27

Goma 16.72 15.6 1.1 15.6 212 quadrangular 1 2 4.86 1 2 6.56 CO 0.50

Goma 9.35 8.83 1.1 8.8 60.53 spherical 0 0 0 1 1 4.675 CU 0.30

Goma 14.5 13.95 1.0 14.0 154 spherical 0 0 0 1 1 6.79 GI 0.33

Goma 14.63 13.13 1.1 13.1 129.55 spherical 0 0 0 1 1 6.97 GI 0.31

Goma 20.31 18.43 1.1 18.4 304.94 polygon 6 3.36 9 1 0 7.76 SA 0.53

Goma 24.12 19.79 1.2 19.8 368.42 ovoid 0 4 1 1 2 12 CO 0.75

Goma 7.38 5.08 1.5 5.1 32.32 ovoid 1 0 0 1 0 3.65 TR 0.72

Goma 29.31 20.18 1.5 20.2 463 ovoid 1 0 0 1 1 20.86 GI 0.45

Goma 17.47 16.82 1.0 16.8 215 ovoid 1 0 0 1 0 8.5 GI 0.72

Goma 11.99 11.99 1.0 12.0 106 spherical 1 3.2 1 1 0 4.84 GI 0.30

Goma 11.57 9.93 1.2 9.9 90.26 spherical 0 0 0 1 0 2.95 GI 0.25

Goma 14.06 11.38 1.2 11.4 129.31 ovoid 0 0 0 1 3 9.76 CO 0.52

Goma 30.68 21.71 1.4 21.7 493.67 ovoid 0 0 0 1 2 10 CO 0.59

Goma 11.33 11.33 1.0 11.0 103.43 spherical 0 0 0 1 0 5.5 CU 0.34

Goma 16.15 15.79 1.0 15.8 201.29 spherical 0 0 0 1 1 8 GI 0.33

Goma 30.27 25.27 1.2 25.3 568.1 triangular 0 0 0 1 2 19 CO 0.65

Goma 5.32 5.12 1.0 5.1 20.16 polygon 8 0 0 3 0 3.81 PI 0.38

Goma 5.32 5.12 1.0 5.1 20.16 polygon 8 0 0 3 0 3.81 PI 0.38

Goma 5.32 5.12 1.0 5.1 20.16 polygon 8 0 0 3 0 3.81 PI 0.38

Goma 5.32 5.12 1.0 5.1 20.16 polygon 8 0 0 3 0 3.81 PI 0.38

Goma 5.32 5.12 1.0 5.1 20.16 polygon 8 0 0 3 0 3.81 PI 0.38

Goma 5.32 5.12 1.0 5.1 20.16 polygon 8 0 0 3 0 3.81 PI 0.38

Goma 5.32 5.12 1.0 5.1 20.16 polygon 8 0 0 3 0 3.81 PI 0.38

Goma 5.32 5.12 1.0 5.1 20.16 polygon 8 0 0 3 0 3.81 PI 0.38

Goma 5.32 5.12 1.0 5.1 20.16 polygon 8 0 0 3 0 3.81 PI 0.38

Goma 5.32 5.12 1.0 5.1 20.16 polygon 8 0 0 3 0 3.81 PI 0.38

Goma 5.32 5.12 1.0 5.1 20.16 polygon 8 0 0 3 0 3.81 PI 0.38

Goma 5.32 5.12 1.0 5.1 20.16 polygon 8 0 0 3 0 3.81 PI 0.38

Goma 5.32 5.12 1.0 5.1 20.16 polygon 8 0 0 3 0 3.81 PI 0.38

Goma 5.32 5.12 1.0 5.1 20.16 polygon 8 0 0 3 0 3.81 PI 0.38

Goma 5.32 5.12 1.0 5.1 20.16 polygon 8 0 0 3 0 3.81 PI 0.38

Goma 5.32 5.12 1.0 5.1 20.16 polygon 8 0 0 3 0 3.81 PI 0.38

Goma 5.32 5.12 1.0 5.1 20.16 polygon 8 0 0 3 0 3.81 PI 0.38

234
Goma 5.32 5.12 1.0 5.1 20.16 polygon 8 0 0 3 0 3.81 PI 0.38

Goma 5.32 5.12 1.0 5.1 20.16 polygon 8 0 0 3 0 3.81 PI 0.38

Goma 5.32 5.12 1.0 5.1 20.16 polygon 8 0 0 3 0 3.81 PI 0.38

Goma 5.32 5.12 1.0 5.1 20.16 polygon 8 0 0 3 0 3.81 PI 0.38

Goma 5.32 5.12 1.0 5.1 20.16 polygon 8 0 0 3 0 3.81 PI 0.38

Goma 5.32 5.12 1.0 5.1 20.16 polygon 8 0 0 3 0 3.81 PI 0.38

Goma 5.32 5.12 1.0 5.1 20.16 polygon 8 0 0 3 0 3.81 PI 0.38

Goma 5.32 5.12 1.0 5.1 20.16 polygon 8 0 0 3 0 3.81 PI 0.38

Goma 5.32 5.12 1.0 5.1 20.16 polygon 8 0 0 3 0 3.81 PI 0.38

Goma 5.32 5.12 1.0 5.1 20.16 polygon 8 0 0 3 0 3.81 PI 0.38

Goma 5.32 5.12 1.0 5.1 20.16 polygon 8 0 0 3 0 3.81 PI 0.38

Goma 5.32 5.12 1.0 5.1 20.16 polygon 8 0 0 3 0 3.81 PI 0.38

Goma 5.32 5.12 1.0 5.1 20.16 polygon 8 0 0 3 0 3.81 PI 0.38

Goma 5.32 5.12 1.0 5.1 20.16 polygon 8 0 0 3 0 3.81 PI 0.38

Goma 5.32 5.12 1.0 5.1 20.16 polygon 8 0 0 3 0 3.81 PI 0.38

Goma 5.32 5.12 1.0 5.1 20.16 polygon 8 0 0 3 0 3.81 PI 0.38

Goma 5.32 5.12 1.0 5.1 20.16 polygon 8 0 0 3 0 3.81 PI 0.38

Goma 5.32 5.12 1.0 5.1 20.16 polygon 8 0 0 3 0 3.81 PI 0.38

Goma 5.32 5.12 1.0 5.1 20.16 polygon 8 0 0 3 0 3.81 PI 0.38

Goma 5.32 5.12 1.0 5.1 20.16 polygon 8 0 0 3 0 3.81 PI 0.38

Goma 5.32 5.12 1.0 5.1 20.16 polygon 8 0 0 3 0 3.81 PI 0.38

Goma 5.32 5.12 1.0 5.1 20.16 polygon 8 0 0 3 0 3.81 PI 0.38

Goma 5.32 5.12 1.0 5.1 20.16 polygon 8 0 0 3 0 3.81 PI 0.38

Goma 5.32 5.12 1.0 5.1 20.16 polygon 8 0 0 3 0 3.81 PI 0.38

Goma 5.32 5.12 1.0 5.1 20.16 polygon 8 0 0 3 0 3.81 PI 0.38

Goma 5.32 5.12 1.0 5.1 20.16 polygon 8 0 0 3 0 3.81 PI 0.38

Goma 5.32 5.12 1.0 5.1 20.16 polygon 8 0 0 3 0 3.81 PI 0.38

Goma 5.32 5.12 1.0 5.1 20.16 polygon 8 0 0 3 0 3.81 PI 0.38

Goma 5.32 5.12 1.0 5.1 20.16 polygon 8 0 0 3 0 3.81 PI 0.38

Goma 5.32 5.12 1.0 5.1 20.16 polygon 8 0 0 3 0 3.81 PI 0.38

Goma 5.32 5.12 1.0 5.1 20.16 polygon 8 0 0 3 0 3.81 PI 0.38

Goma 5.32 5.12 1.0 5.1 20.16 polygon 8 0 0 3 0 3.81 PI 0.38

Goma 5.32 5.12 1.0 5.1 20.16 polygon 8 0 0 3 0 3.81 PI 0.38

Goma 5.32 5.12 1.0 5.1 20.16 polygon 8 0 0 3 0 3.81 PI 0.38

Goma 5.32 5.12 1.0 5.1 20.16 polygon 8 0 0 3 0 3.81 PI 0.38

Goma 5.32 5.12 1.0 5.1 20.16 polygon 8 0 0 3 0 3.81 PI 0.38

Goma 5.32 5.12 1.0 5.1 20.16 polygon 8 0 0 3 0 3.81 PI 0.38

Goma 5.32 5.12 1.0 5.1 20.16 polygon 8 0 0 3 0 3.81 PI 0.38

Goma 5.32 5.12 1.0 5.1 20.16 polygon 8 0 0 3 0 3.81 PI 0.38

Goma 5.32 5.12 1.0 5.1 20.16 polygon 8 0 0 3 0 3.81 PI 0.38

Goma 5.32 5.12 1.0 5.1 20.16 polygon 8 0 0 3 0 3.81 PI 0.38

Goma 5.32 5.12 1.0 5.1 20.16 polygon 8 0 0 3 0 3.81 PI 0.38

Goma 5.32 5.12 1.0 5.1 20.16 polygon 8 0 0 3 0 3.81 PI 0.38

Goma 5.32 5.12 1.0 5.1 20.16 polygon 8 0 0 3 0 3.81 PI 0.38

Goma 5.32 5.12 1.0 5.1 20.16 polygon 8 0 0 3 0 3.81 PI 0.38

Goma 5.32 5.12 1.0 5.1 20.16 polygon 8 0 0 3 0 3.81 PI 0.38

235
Goma 5.32 5.12 1.0 5.1 20.16 polygon 8 0 0 3 0 3.81 PI 0.38

Goma 5.32 5.12 1.0 5.1 20.16 polygon 8 0 0 3 0 3.81 PI 0.38

Goma 5.32 5.12 1.0 5.1 20.16 polygon 8 0 0 3 0 3.81 PI 0.38

Goma 5.32 5.12 1.0 5.1 20.16 polygon 8 0 0 3 0 3.81 PI 0.38

Goma 5.32 5.12 1.0 5.1 20.16 polygon 8 0 0 3 0 3.81 PI 0.38

Goma 20.17 20.17 1.0 20.0 312.77 spherical 0 0 0 1 1 6.3 CU 0.32

Goma 23.46 11.8 2.0 11.8 213.82 elongate ovoid 0 0 0 1 2 9.54 CO 0.52

Goma 10.44 10.44 1.0 10.2 80.19 spherical 0 0 0 1 1 9.76 CU 0.38

Goma 9.04 8.36 1.1 8.4 60.76 spherical 0 0 0 1 1 4.5 CU 0.32

Goma 11.05 10.06 1.1 10.1 84.6 spherical 0 0 0 1 1 5 GI 0.30

Goma 6.74 4.71 1.4 4.7 23.52 prolate 0 0 0 1 1 1.95 XY 0.23

Goma 14.35 12.35 1.2 12.4 180 polygon 6 0 0 1 1 5.87 SA 0.62

Goma 21.41 15.77 1.4 15.8 261 ovoid 0 0 0 1 1 14.55 GI 0.37

Goma 9.42 7.68 1.2 7.7 63.58 hemispherical 1 0 0 2 1 4.89 GI 0.51

Goma 8.63 6.06 1.4 6.1 40.83 hemispherical 1 0 0 2 0 4.5 GI 0.46

Goma 15.43 12.53 1.2 12.5 166.28 polygon 4 0 0 1 0 6.75 SA 0.47

Goma 7.99 7.17 1.1 7.2 46.48 ovoid 0 0 0 1 0 4.2 TR 0.62

Goma 12.46 12.46 1.0 11.7 111.51 spherical 0 0 0 1 2 6.81 CU 0.62

Goma 25 18 1.4 18.0 372 ovoid 0 0 0 1 0 13 CO 0.34

Rubra 9.84 12.34 0.8 12.3 95.74 hemispherical 3 0 0 2 0 5.08 GI 0.47

Rubra 5.29 4.37 1.2 4.4 16.86 spherical 0 0 0 1 0 1.79 ER 0.27

Rubra 7.39 5.59 1.3 5.6 30.21 ovoid 0 0 0 1 0 1.87 TR 0.71

Rubra 9.95 9.95 1.0 9.7 72.27 spherical 0 0 0 1 0 4.17 CU 0.36

Rubra 6.9 5.55 1.2 5.6 28.27 prolate 0 0 0 1 0 2.89 TR 0.31

Rubra 7.97 6.41 1.2 6.4 43.34 elongate 2 0 0 1 0 4.61 TR 0.26

conovoid
Rubra 16.52 12.95 1.3 13.0 182.78 polygon 7 0 0 1 0 7.23 SA 0.64

Dorry 8.81 7.6 1.2 7.6 51.33 ovoid 0 0 0 1 0 3.34 TR 0.58

Dorry 8.61 6.05 1.4 6.1 38.2 prolate 0 0 0 1 0 3.6 TR 0.34

Dorry 5.84 5.22 1.1 5.2 26.04 spherical 0 0 0 1 0 2.07 NA 0.32

Dorry 6.96 6.48 1.1 6.5 35.5 spherical 0 0 0 1 1 3.48 CU 0.28

Venus 9.43 7.9 1.2 7.9 65.51 spherical 1 0 0 1 0 4.3 GI 0.33

Venus 9.93 9.01 1.1 9.0 73.23 spherical 0 0 0 1 0 6.59 GI 0.32

Venus 17.91 14.01 1.3 14.0 185.85 ovoid 0 0 0 1 1 11.17 GI 0.46

Venus 8.4 7.79 1.1 7.8 53.82 spherical 0 0 0 1 0 3.38 CU 0.25

Venus 4.53 4.12 1.1 4.1 15.04 hemispherical 1 0 0 1 0 1.91 XY 0.23

Venus 15.97 11.92 1.3 11.9 134.03 pyriform 0 0 0 1 1 8.2 CO 0.44

Venus 7.48 6.16 1.2 6.2 33.17 prolate 1 0.5 2 1 1 3.18 PA 0.33

Venus 6.04 4.22 1.4 4.2 23.28 prolate 0 0 0 1 0 2.11 PA 0.31

Venus 12.17 10.44 1.2 10.4 93.68 prolate 0 0 0 1 0 5.64 CO 0.48

Venus 15.61 12.87 1.2 12.9 150.5 prolate 0 0.7 2 1 0 8.2 CO 0.44

Venus 26.95 20.72 1.3 20.7 435 ovoid 0 0 0 1 2 20 CO 0.58

Venus 9.56 8.69 1.1 8.7 65.62 hemispherical 0 0 0 2 0 5.49 GI 0.47

Venus 10.79 10.72 1.0 10.7 93.61 hemispherical 0 0 0 2 0 6.04 GI 0.42

Venus 14.25 12.7 1.1 12.7 143.15 spherical 0 0 0 1 1 8 GI 0.32

Venus 8.43 7.16 1.2 7.2 44.74 spherical 0 0 0 1 1 4.15 CU 0.27

Venus 4.89 3.26 1.5 3.3 12.52 prolate 0 0 0 1 0 1.77 PA 0.42

236
Brutus 6.41 5.65 1.1 5.7 34.96 hemispherical 2 0 0 2 0 3.32 GI 0.40

Brutus 6.02 5.06 1.2 5.1 24.28 hemispherical 2 0 0 2 0 2.87 SG 0.38

Brutus 4.73 4.35 1.1 4.4 17.12 ovoid 0 0 0 1 0 2.35 TR 0.28

Brutus 21.56 16.95 1.3 17.0 181 ovoid 0 0 0 1 0 15.88 GI 0.43

hector 8.19 8.19 1.0 8.0 50.66 spherical 0 1.02 1 1 0 3.38 CU 0.47

hector 5.95 5.39 1.1 5.4 25.88 oblate conovoid 2 0 0 2 0 2.61 SG 0.57

Lefkas 6.48 5.63 1.2 5.6 27.02 spherical 0 0 0 1 0 3.24 NA 0.28

Lefkas 4.35 4.13 1.1 4.5 18.23 spherical 1 0 0 1 0 2.175 NA 0.27

Lefkas 19.67 18.32 1.1 18.3 276.13 ovoid 1 0 0 1 3 11.29 CO 0.41

Lefkas 3.61 3.61 1.0 3.6 9.81 spherical 0 0 0 1 0 1.805 ER 0.62

Lefkas 4.53 3.35 1.4 3.4 11.13 spherical 0 0 0 1 1 1.65 ER 0.34

Lefkas 5.18 5.18 1.0 4.4 18.33 spherical 0 0 0 1 0 1.65 ER 0.41

Lefkas 3.98 3.98 1.0 3.9 14.39 spherical 0 0 0 1 0 1.99 ER 0.57

Lefkas 7.48 4.4 1.7 4.4 29.47 ovoid 0 0 0 1 0 3.74 TR 0.70

Agathe 12.37 7.81 1.6 7.8 77.7 ovoid 0 0 0 1 0 6.185 TR 0.39

Agathe 7.82 6.97 1.1 7.0 42.71 spherical 2 0 0 1 0 3.88 GI 0.30

Agathe 7.85 6.65 1.2 6.7 39.78 spherical 1 0 0 1 1 2.31 CU 0.26

Agathe 5.8 5.25 1.1 5.3 24.04 spherical 1 0 0 1 1 2.9 CU 0.31

Agathe 8.23 6.03 1.4 6.0 37.76 prolate 0 0 0 1 0 3.62 TR 0.34

Agathe 4.58 4.22 1.1 4.2 15.2 spherical 0 0 0 1 0 2.29 NA 0.31

Clyde 12.65 10.84 1.2 10.8 106 spherical 2 0 0 1 0 4.75 GI 0.49

Clyde 6.01 4.5 1.3 4.5 21.7 spherical 0 0 0 1 2 3.005 CU 0.50

Clyde 13.31 12.49 1.1 12.5 133 polygon 8 1.7 2 3 0 4.93 SA 0.78

Tina 5.34 4.72 1.1 4.7 20.45 oblate conovoid 2 0 0 2 0 2.39 SG 0.65

Tina 4.92 3.98 1.2 4.0 15.51 hemispherical 1 0 0 1 0 1.85 SG 0.25

Tina 8.92 8.64 1.0 8.6 68.1 spherical 0 0 0 1 0 4.32 GI 0.27

Oreste 6.5 4.81 1.4 4.8 27.18 ovoid 1 0 0 1 0 2.44 TR 0.65

Oreste 6.16 5 1.2 5.0 25.54 oblate conovoid 2 0 0 2 0 2.47 SG 0.57

Vanessa 8.65 7.85 1.1 7.9 48.29 spherical 0 0 0 1 0 3.94 CU 0.25

Vanessa 9.37 8.2 1.1 8.2 62.79 prolate 0 0 0 1 0 3.65 CO 0.35

Appendix table 13: Coefficients of the statistical models.

Model Term Estimate Std. Err. Z value P

Tests of effect of age and sex on microremain numbers
Phytolith Negative binomial Intercept 3.969 0.160 24.790 1.1398e-135
Age 0.002 0.0005 3.833 1.2616e-04
Sex -0.027 0.157 -0.170 8.6469e-01
Starch Negative binomial Intercept 3.009 0.426 7.052 1.7575e-12
Age 0.003 0.001 2.661 7.7805e-03
Sex -2.569 0.437 -5.873 4.2665e-09
Unsilicified remains Intercept 2.210 0.202 10.904 1.0978e-27
Negative binomial Age 0.001 0.0006 3.093 1.9775e-03
Sex -0.048 0.199 -0.245 8.0594e-01
Tests of effect of consumption frequency on microremain numbers

237
Phytolith Poisson model Intercept -0.231 0.876 -0.263 0.791
z.min 1.707 0.680 2.509 0.012
z.age 3.612 2.075 1.740 0.081
sex -0.801 0.934 -0.858 0.390
Starch logistic regression model Intercept -14.218 0.870 -6.325 6.4911e-60
z.min 0.591 0.505 1.169 2.4224e-01
z.age 0.489 0.442 1.105 2.6885e-01
sex -1.266 0.996 -1.271 2.0372e-01

Appendix table 14: Variable importance in phytolith and starch classification random forest.

Phytolith model Starch model

Variable Importance Variable Importance
Length 100 Area 100
Spine number 75.301 Length 75.8434
Spine ang 74.109 Width 67.5876
LW Ratio 43.996 Dist 61.1718
Spine length 42.854 Facets 60.4963
Area 29.581 LW Ratio 56.2298
Width 22.056 Type 55.9587
Irregul 10.236 Lam 35.8372
Spherical 6.667 Spherical 31.833
Angularpoint 6.575 Prolate 8.2554
Polygon 4.590 Ovoid 7.157
Ovoid 1.663 Polygon 5.8693
Prolate 1.620 Hemispherical 4.9279
Triangular 1.447 Oblate conovoid 4.6926
Elongate 0.440 Striaelen 2.4395
Quadrangular 0.228 Elongate ovoid 2.4011
Facets 0.184 Striae no 2.1051
Conjoined 0.106 Triangular 1.8956
Prolate concave-convex 0.043 Quadrangular 0.9141
Polygon concave 0.042 Pyriform 0.4986

7.3 Chapter five appendix

7.3.1 Comparative data for model

We prepared data from past dental calculus studies for a comparative

analysis (Salazar-García et al., 2013; Henry et al., 2014). This dataset included starch
and phytolith counts from nine other Middle Palaeolithic sites. As other
microremains are not included in previous published studies, we only included
starch and phytoliths in our model. Although our samples were weighed in mg,
weights for all eight sites are not available. Similarly, in the datasets presented in this
paper we treated starches of the same type that occurred as lumps as one starch as

238
accurately counting each starch in a lump is not possible. We collected the most to
date estimated date range for each site and used the median value.

Goyet: this archaeological site comprises several caves near Gesves, in the
Namur Province of Belgium. The cave system has seen several campaigns of
excavation in the 19th and 20th century. Early explorers found hominin remains
(Goyet VIII) in 1868 in the largest of the caves. Dupont found the studied mandible
in the second of five fauna-rich levels (Dupont, 1872; Toussaint, 2006). Originally, the
fossil was thought to be modern human due to its stratigraphic proximity to
Aurignacian artefacts, but this has been re-evaluated and it now is accepted to be a
Neanderthal (Rougier et al., 2012, 2014). In addition, in the Aurignacian phase there
is an upper Magdalenian level dated to 13 ka (Toussaint, 2006). Mixing is present in
all levels and its date was long ambiguous but this has recently been re-evaluated as
dating to 44-45.5 ka cal BP (Rougier et al., 2014). This date places the hominin in a
transitional period. Regional vegetation reconstructions suggest the surrounding
environment was generally tundra-steppe.

La Chapelle-aux-Saints: this Middle Palaeolithic site is located in the Corrèze

region of southern France. Researchers have excavated La Chapelle-aux-Saints since
1905, and this has recovered evidence of Mousterian sediments and a complete
Neanderthal in 1908. The chronological history of this site has been studied with
electron spin resonance (ESR), suggesting dates of 56 ka or 47 ka depending on the
radiation uptake model used (Grün and Stringer, 1991). The ESR may suggest the
remains belongs to the warm parts of MIS 3, but this contradicts correlation with the
Combe-Grenal sequence which would put the remains at the end of MIS 4 and
beginning of MIS 3. The associated fauna profile is predominately reindeer (Rangifer
tarandus), with some bovines (Bos/Bison sp.), horse (Equus sp.), ibex, wolf (Canis
lupus), fox (Canis vulpes), Rhinoceros, cave hyena (Crocuta spelaeaus), boar (Sus scrofa)
and marmot (Arctomys sp.) (Boule, 1911; Bouyssonie et al., 1913). The fauna is clearly
a cold phase profile indicating a date during the late MIS 4 (Mellars, 1986). In
addition, fauna shows the surrounding environment was a cold open biome.

La Ferrassie: this site is located in the Vézère Valley, in the Dordogne region
of France. La Ferrassie is a large deep cave with an adjoining long rock-shelter and
small rock-shelter. The site has a plethora of levels of different periods in various
sections of the cave. Mousterian levels below the long rock-shelter produced remains
of six Neanderthals in excavations during 1909 and 1921. The bison, auroch and red
deer that dominate the Mousterian fauna imply a moderate temperate environment.

239
These fauna suggest tree cover and a closed, forested environment (Capitan and
Peyrony, 1912a; b; c; Guérin et al., 2015). Mousterian deposits at La Ferrassie has
been recently dated with OSL and radiocarbon dating, suggesting that the
Neanderthal remains La Ferrassie 1 is most likely 39 ±5 ka and 2 skeletons 43 ±3 ka
(Guérin et al., 2015).

La Quina: La Quina is a series of rock shelters in the Charente region of

Central France. Remains used in this study were found in 1911 in one of two sub-
sections of Station Amont, a deposit extending below the upper rock shelter base.
This deposit was studied over the course of several excavations. Excavations
revealed Mousterian remains, faunal debris and the remains of many Neanderthals
(Henri-Martin, 1961). The upper deposits of the sequence at Station Amont are
considered to date to 48-43 ka. This, combined with cold phase fauna, indicates a
date for the fossil of MIS 4, probably 71-57 ka (Debénath and Jelinek, 1998). Fauna
found was mostly bovines, horse and reindeer, with few other species represented
(Debénath and Jelinek, 1998). These faunas also suggest a cold and dry environment
that was devoid of trees.

Malarnaud: this site is a cave in the Ariège region of Southern France. There
has been scientific interest in the cave since 1883. Deposits dated to Mousterian,
Aurignacian and Magdalenian have been found onsite. Investigators found a
juvenile Neanderthal mandible during 1888 in the lower of two layers in a side
chamber of this cave complex. However, it is possible that the mandible was moved
by carnivores in this chamber as it is removed from much the archaeological
material. Unfortunately, the site has not been radiometrically dated. Faunal profiles
indicate the mandible dates to Riss-Würm interglacial, 130-117 ka or the beginning of
the Würm, 100-50 ka. Fauna in the layer of the mandibles include cave lion (Panthera
leo), cave hyena, fox, and wolf, mammoth and rhinoceros (Rhinocerotidae) (Boule,
1889; Filhol, 1889). This fauna is suggestive of tree cover in the early glacial warm or
transitional phase, and thus we classify the environment as of mixed openness.

Spy: this archaeological site is located in Jemeppe-sur-Sambre, Namur in

Belgium. The site was excavated from 1879 onwards, and the Neanderthal remains
were found in a bone rich layer. Later excavations have clarified the stratigraphy of
the cave. Faunal profiles from excavation of this layer have suggested an intensely
cold climate (Otte, 1979). Some studies found misclassified Neanderthal remains in
faunal bags (Crevecoeur et al., 2010). These teeth were directly radiocarbon dated to
about 36 ka (Semal et al., 2009). De Puydt and Lohest recovered fauna from this

240
level, including horse and hyena, with some mammoth, wholly rhinoceros, reindeer,
red deer, aurochs, cave bear, cave lion, wolf, wolverine (Gulo gulo) and badger (Meles
meles). However, palaeoenvironment reconstructions may be questioned due to the
poor stratigraphic integrity of this layer (de Puydt and Lohest, 1887). The direct data
of the hominin remains firmly places the occupation in a cold phase when dry tree
landscapes dominated much of Europe. We consider the environment as open for
our model.

Kůlna Cave: this Middle Palaeolithic site is located in the Moravian Karst, in
the eastern part of the Czech Republic in Central Europe. The cave saw first
investigations in 1880 when stone tools and bones of extinct animals were noticed
(Sroubek et al., 2001). Karel Valoch conducted the first modern archaeological
investigation in 1961 and 1976. He identified 14 sedimentary complexes covering the
last interglacial to the Holocene. Neanderthal remains were found in strata 7a and 7c
of but specimens in this study come from stratum 7a only. Radiocarbon dating has
suggested a data of >45 ka BP 14C, and electron spin resonance on layer 7a shows it
dates to 50 ± 5 ka BP (Rink et al., 1996). The character of the fauna from this layer
matches this age (Rink et al., 1996). Layer 7a contained reindeer, with mammoth and
a few elk, the presence of reindeer clearly indicate cold conditions of central Europe
in the MIS 3 (Valoch, 1970).

Shanidar Cave: This site is located in the Zagros Mountains in Northwest

Iraq. Solecki and colleagues excavated the cave between 1952 and 1957. Excavators
described four archaeological strata (A, B, C and D). The Shanidar III fossils were
found in Mousterian level D (Solecki, 1960). A radiocarbon date near the Shanidar I
fossil indicates that Shanidar III is >46 ka BP, possibly as old as 50 ka BP (Solecki,
1960). Goat (Capra sp.) and sheep (Ovis sp.) dominate fauna found on site. This
reflects the local mountainous topography (Perkins, 1964; Evins, 1982). Pollen
analysis indicated the presence of date palms (Phoenix dactylifera), walnuts (Juglans
sp.), chestnuts (Castanea sp.), oaks (Quercus spp.) and herbs (Solecki, 1961; Leroi-
Gourhan, 1968, 1969, 1975). These plant taxa indicate a mild moist environment with
at least some level of tree cover. For our model, we classified this habitat as closed.

7.3.2 Reference collection

Microremain identification was based on a reference collection of modern

plant samples, including >2,000 global species. Our reference collection has extensive

241
coverage of edible western Eurasian species. From these species, we identified over
54 species that produced starches, and thus that might be represented in our samples
(Appendix table 21). More information is available for phytoliths produced by
different taxa so; we instead identified phytoliths using available literature including
PhyCore database (Albert et al., 2016). We did not make a reference collection for
unsilicified plant microremains, as its unclear if these microremains are diagnostic,
nor do we currently have a sufficient reference collection for identifying this types of
microremains (Power et al., 2015b).

7.3.3 Classification of microremain taxa

We identified microremains to plant taxon, usually at the family or tribe level.

When this was not possible, we assigned microremains to a type based on shared
diagnostic morphology that indicates that the morphology likely represents a single
plant taxon. We then used the summed number of types to derive a metric of
breadth of plant use.

7.3.4 Microremain results

See following tables.

242
Appendix table 15: Total recovered microremains from Vindija Cave Neanderthal and control samples.

Starches Phytoliths

Identifier Specimen Species Tooth Wt

unknown phyto hair

Long-cell multi-cell
partially disrupted

Globular sinuate?

Parallelepipedal

Multicellular
polyhedrons
Brachiform
Bulliform
Long-cell
Pos/Dmg
Type 10

Type 11

Type 12

Type 13

Type 14

Type 15

Rondel
Type 1

Type 2

Type 3

Type 4

Type 5

Type 6

Type 7

Type 8

Type 9

Hair
Vja-12-13 12.1/229 Neanderthal URM2 0.39 3 3 5 1 3 1 8 3 3 1 4 3 1 2

Vja-12-14 12.2/286 Neanderthal LRI2 0.05 1 1 1 2 1 1

Vja-12-16 12.4/ 290 Neanderthal URI1 0.05 2

Vja-12-17 12.5/287 Neanderthal URC 0.05 1 1 1

Vja-12-18 12.6/288 Neanderthal LLC 0.02 1 1 1 1

Vja-12-19 12.7/201 Neanderthal LLI2 0.89 4 4 1 2 1 1 8 3 3 5 2 3

Vja-12-20 11.39/206 Neanderthal LRC 0.45 1 1 3 2

Vja-12-21b 11.39/206 Neanderthal LRM1 0.41 1 1 1 1 4 1 1

Vja-12-21a 11.39/206 Neanderthal LRM1 0.50 1 1 1 1

Vja-12-24 11.45/231 Neanderthal LLM3 0.67 1 1 1 4 4 1 3 1

Vja-12-26 11.46/259 Neanderthal ULM2 0.87 3 2 1 1 5 3

Vja-12-51 11.40+11.40a – Vi76/226+265 Neanderthal LLM1 0.19 1 1 1 1 1 1 2 1

Vja-12-54 11.40+11.40a – Vi76/226+265 Neanderthal LLM1 0.05 1 1

Vja-12-55 11.40+11.40a – Vi76/226+265 Neanderthal LLM1 0.09

Vindija fauna calculus samples

Vja-12-28 1639/car Vi-87 Panthera C 0.63

Vja-12-29 555/car 78 Canis 0.79 1

Vja-12-30 735/car Vi-83 Canis 0.53 7 3 1 1

Vja-12-31 335/car Vi-78 Canis P1 0.10 1 1 1

Vja-12-34 637/car Vi-76 Canis 0.40

Vja-12-35 714/car Vi-1976 Canis 0.58 1

Vja-12-37 2/car Vi-76 Ursus spelaeus M2 1.04

Vja-12-38 nova 3/car Vi-76 U.spelaeus M2 0.46 1 1 8

Vja-12-45 nova 1/car Vi-76 U.spelaeus M2 1.248 2

Vja-12-46 U.spelaeus M2 0.88 1 4 1 1 3 1 3

Vja-12-47 U.spelaeus M2 0.18 1 2 1 2 5

Vja-12-48 U.spelaeus M2 0.30 1 3

Vja-12-53 599 Vi-78 Canis 0.43

243
Controls

Vja-12-43 adhesive used to hold teeth, sampled on Vi-11.39 0.796 1 1 2

Vja-12-44 adhesive used to hold teeth, sampled on Vi-11.40 0.54 1 7 11 4 30

Phytoliths (continued) Calcium oxalates Spores Pollens Spicul Hai Unsilicified plant
e r microremains
Identifier

Ellipsoidal, single-walled

Total starch & phytoliths

Total starch & phytoliths

Very small round dark

Unknown unsilicified
Yellow pollen/spore?

Monocot unsilicified
Pollen (Betulaceae)

Cellouse type fibre

Menhinick's index
Indet. animal cells
Ellipsoid rugulate

Unknown spicule

Grass unsilicified
Vascular bundle
Unknown spore

Tube of spheres
Indet. phytolith

Clear fusiform

Unidentified

Pollen indet.
Irreg oxalate

Nematode??
Epidermis

spherulite
Cluster
Styloid

Algae?
Prism

spore

types
Fibre
Plate

Hair
Vja-12-13 2 1 2 2 1 37 13 2.
1
Vja-12-14 1 3 1 5 5 2.
2
Vja-12-16 1 1 1 2 1 0.
7
Vja-12-17 2 2 1 2 3 3 1.
7
Vja-12-18 3 1 1 2 3 3 1.
7
Vja-12-19 2 1 1 2 2 1 1 1 1 32 13 2.
3
Vja-12-20 7 4 1.
5
Vja-12-21b 1 2 3 1 1 10 7 2.
2
Vja-12-21a 4 4 2
Vja-12-24 1 1 1 1 1 3 17 9 2.
2
Vja-12-26 1 3 1 1 2 1 1 1 1 1 3 15 6 1.
5
Vja-12-51 3 1 2 3 2 7 6 2.
3
Vja-12-54 2 2 2 1.
4
Vja-12-55 1 2 1
Vindija fauna calculus samples
Vja-12-28 1 2
Vja-12-29 1 1
Vja-12-30 4 4 1 12 4 1.
2
Vja-12-31 3 3 1.
7

244
Vja-12-34 1 1 1
Vja-12-35 1 1 1 1
Vja-12-37 1
Vja-12-38 1 1 2 1 1 1 2 12 5 1.
4
Vja-12-45 1 2 1 1 3 2 1.
2
Vja-12-46 1 1 1 3 14 7 1.
9
Vja-12-47 1 1 4 3 11 5 1.
5
Vja-12-48 1 1 1 5 4 2 1
Vja-12-53 2 3 3
Controls
Vja-12-43 2 2
Vja-12-44 23 4
Starch key
Type 1 Moderate size, spherical-subspherical, with thick lamellae, some show yellow colouration, diameter is 10-22 µm.
Type 2 Large circular-subcircular in 2D, spherical-lenticular-subspherical 3D, diameter is 20->µm.
Type 3 Small round, constrained facets may be present, diameter is <10 µm.
Type 4 Sub-polyhedral, 2 or more facets but more of surface is not covered by facets, facets often are less sharply defined, no lamellae.
Type 5 Slightly eccentric starch.
Type 6 Faceted, generic type.
Type 7 Ovoid starch, with or without surface features, some have damaged central cavity but this is not a classification trait.
Type 8 Triangular-elliptical, may have central fissure, other surface features can include lamellae.
Type 9 Very eccentric and partially disrupted starch.
Type 10 Lenticular or subelliptical in 3D, equatorial groove may be visible, some show signs of gelatinisation, distinguished from type 8 by poorly defined longitude crack.
Type 11 Small oval or slight ovoid, subspherical (5-10 µm), 1-2 facets may be apparent, little surface features but a central aperture may be present.
Type 12 Large ovoid, routinely eccentric, often with lamellae, diameter is >40 µm.
Type 13 Large spherical/subspherical
Type 14 Polyhedral, distinct facets surface on ≥50 %, no lamellae present.
Type 15 Very small polyhedral, highly facets surface on ≥50 %, no lamellae present.

245
Appendix table 16: Total recovered microremains from Grotta Guattari Neanderthal and control samples. Specimen column uses Circeo numbering.

Starches Phytoliths Spores Other microremains

Prism & sub calcium oxalate

Bulliform Parallelepipedal

Multicellular indet.
Spheroid granulate

fusiform (boletoid)

Pollen with airsacs

cf Pteridium spore
Cylindroid psilate

Indeterm. pollen
Parallelepipedal

Triporate pollen

Indet. plant cell

Vascular budle

Menhinick's
Faunal Hair
Indet. spore

Stellate hair
Nigrospora
Dmg/indet.

Mesophyll

Grass cells
Long Cell
Specimen

Trichome
Identifier

Cystolith
Type 11

Diatom
Type 2.

Rondel
Type 1

Type 3

Type 4

Type 5

Type 6

Type 7

Type 8

Type 9
Tooth

Other

hair
Wt

Neanderthal samples

GTN1 2 RLM3 0.65 1 1 1 2 1 2.1

GTN2 3 RLM1 0.87 1 1 1 7 6 5 4 1 2 1 1 1 2 1 2 1 2 1 1 2.5

GTN3 3 LLI2 0.65 2 1 1 2 4 5 3 2 2 2 2 2 1 1 1 1 4 2 1 4 1 1 8 2.6

GTN4 3 RLI2/M1 0.26 1 1 2 1 1 2 1 4.0

GTN5 3 LLM1/PM 0.29 1 1 1 1 2 3.7

1

Guattari Control samples

2a 2 Wash n/a 7 4

2b 2 Wash n/a 1 1

2c 2 Wash n/a 1 1 1 1

2d 2 Wash n/a 1 7 1 1

2e 2 Adhesive 1.30 2 1 1 1 0.3

0
4
2g 2 Bone dust 0.34

3a 3 Filler #### 4 0.1

3b 3 Adhesive n/a 2 2 1
residue
Starch key

Type 1: Polyhedral with centric extinction cross oval with no fissures, cross arms are clear and straight, diameter is 17 µm.

Type 2: Unknown shape, partially disrupted (semi gelatinised) eccentric starch, diameter is 90 µm.

Type 3: Spherical starch, typically with cross arms that are clear and straight or near straight. No discernible surface features. Diameter is 6-9 µm.

Type 4: Sub-polyhedral, cross arms are faint and straight.

Type 5: lenticular, cross arms clear and straight. Faint lamellae present. Diameter is 17 µm. (Possible Triticeae).

Type 6: large sub polyhedral. 15 µm or above.

246
Type 7: Very small polyhedral, no lamellae or fissures (Possible Avena or bogbean).

Type 8: slightly eccentric.

Type 9: Highly eccentric.

Type 10: Very small starch with centric cross.

Type 11: with think lamellae, diameter is 10-20 µm.

247
Appendix table 17: Total recovered microremains from Grotta Fossellone Neanderthal remains and control samples.

. Starches Phytoliths Unsilicified plant microremains

Identifier Specimen Tooth Wt Sampling date

Total starch & phytoliths

Total starch & phytolith

Total numbers/mg
Indet. plant cells
Vascular bundle

Menhinick's/mg
Indet. monocot

Menhinick's
Stellate hair
Grass cell
Type 10

Rondel
Type 2

Type 5

Type 6

Type 7

Xylem
Other

types
Fossellone Neanderthal samples

FON1 Fossellone 3 LLM1 0.07 16th Mar 13 3 1 59.70 3 1 0.58 2.2

FON2 Fossellone 3 LLM2 0.1 16th Mar 13 4 1 1 1 1 1 1 1 9 6 2.00 2.6

Fossellone Control samples

FON3 Fossellone 3 Wash 16th Mar 13 17*

Starch key

Type 1: Polyhedral with centric extinction cross oval with no fissures, cross arms are clear and straight. Diameter is 17 µm.

Type 2: Unknown shape, possible semi gelatinised eccentric starch, diameter is 90 µm.

Type 3: Spherical starch, typically with cross arms that are clear and straight or near straight. No discernible surface features. Diameter is 6-9 µm.

Type 4: Sub-polyhedral, cross arms are faint and straight.

Type 5: lenticular, cross arms clear and straight. Faint lamellae present. Diameter is 17 µm, (Possible Triticeae).

Type 6: large sub polyhedral. 15 µm or above.

Type 7: Very small polyhedral, no lamellae or fissures (Possible Avena sp. or bogbean).

Type 8: slightly eccentric.

Type 9: Highly eccentric.

Type 10: Very small starch with centric cross.

248
Appendix table 18: Total recovered microremains from Sima de las Palomas del Cabezo Gordo Neanderthal and control samples.

Starches Phytoliths Other microremains

Identif Specim Species Tooth Wt.

Total starch & phytoliths/mg

ier en

Total starch & phytoliths

Total starch & phytoliths

Total starch & phytoliths

Degraded glove starch

Menhinick's index/mg
Polyhedral multicell

Menhinick's index
Indet. phytolith
Parallelepiped
Sampling date

Possible algae
Indet. particle

Possible fibre
Dmg/indet

Bulliform

types/mg
Shortcell
Type 1

Type 2

Psilate
Other

types
SP45 SP45 Neanderthal LRP3 0.08 25.July.15 0 0 0

SP54 SP54 Neanderthal LRC 0.10 20.July.15 1 1 1 1 9.8 9.8 3.1

SP83 SP83 Neanderthal LRdm2 0.09 25.July.15 0 0

SP78 SP78 Neanderthal LLP4 n/a 30.June.11 1 1 1 1

SP79 SP79 Neanderthal ULI1 n/a 30.June.11

SP84 SP84 Neanderthal LLM1 n/a 1.July.11 1 1 6 1 2 2 1.4

SP84 SP84 Sediment on SP84 LLM1 n/a 1.July.11

SP1 SP1 Teeth consolidant LRM3 n/a 22.Mar.12 44 44 1 0.2

Packing cotton n/a 1.July.11 1 3 1 2 0 0

SPF1S SPF1S Lagomorph n/a 27.Mar.12 1 2 1 1 4 3 1.5

SPF1D SPF1D Lagomorph 0.33 19.Jul.13 1 0 0 0

SPF2S SPF2S Carnivore P n/a 26.Mar.12 0 0

SPF3 SPF3 Carnivore M n/a 0 0

SPF4 SPF4 Carnivore M n/a 25.Mar.12 1 1 0 0

SPF5D SPF5D Carnivore M n/a 8.Apr.12 1 1 1 1 1

SPF5S SPF5S Carnivore M n/a 5th Apr 12 1 1 1 1

SPF7 SPF7 Horse LRM3 n/a 8th Apr 12 1 1 1 2 2 1.4

Starch key

Type 1 Polyhedral, moderate size, aggregating type, diameter is 8-25 µm.

Type 2 Lenticular.

249
Appendix table 19: Total recovered microremains from Kalamakia Cave Neanderthal remains samples.

Starches Phytoliths Calcium oxalates

Identifier Specimen Type Tooth Wt

Total starch & phytoliths types

Total starch & phytoliths/mg

Total starch & phytoliths

Total starch & phytoliths

Menhinick's index/mg
Menhinick's index
Possible starch
Sampling date

block & sub

Dmg/indet.

types/mg
Irregular

Spicules
Tabular
Rondel
Type 1

Type 2

Indet.

5 face
KAL 3 KAL 3 Neanderthal UlM3 2.87 12.Feb.13 3 1 1 1 4 2 1 1.40 0.70 0.59

KAL 5 KAL 5 Neanderthal URP2 0.05 29.Jan.13 1 14 1 1 1 20 20 4.47

KAL 8 KAL 8 Neanderthal URM2 n/a 26.Jan.13 1 1 1 2 2 1.41

Starch key

Type 1 Lamellae, faint cross on cross polarization.

Type 2 Faceted.

250
 200

150
Total starch and phytoliths

100

50

Appendix fig. 3: Total numbers of starch and phytoliths in each Neanderthals site with reference groups (Twe
forager-horticulturalists from Namibia and Taï Forest Chimpanzees) from Leonard et al., 2015 and Power et
al., 2015.

251
Appendix table 20: Coefficients of statistical models.

Model Term Estimate Std. Err. Z value P

Tests of effect of openness, MET and age on microremain diversity
Random effect Poisson model Intercept 0.918 0.756 1.205 0.228
Openness mixed 1.189 1.259 0.945 0.345

Openness open 1.241 0.500 2.481 0.013

MET -0.464 0.708 -0.656 0.512
Age of fossil specimen -0.432 0.278 -1.553 0.121
Random effect Poisson model Intercept 1.039 0.496 2.094 0.036
with alternative chronology Openness mixed 0.429 1.223 0.351 0.726

Openness open 1.200 0.636 1.886 0.059

MET -0.017 0.445 -0.037 0.970

Alternative age of fossil -0.198 0.241 -0.824 0.410

specimen

Appendix table 21: Western Eurasian economic plants that we identified as starch-rich plants. These plants
are candidate plant food staples.

Family Species Common name

Anacardiaceae Pistacia sp. pistachio
Amaryllidaceae Allium ursinum ramson
Apiaceae Pastinaca sativa wild parsnip
Apiaceae Conopodium majus pignut
Alismataceae Sagittaria sagittifolia arrowhead
Alismataceae Alisma plantago-aquatica water plantain
Araceae Arum maculatum arum
Butomaceae Butomus umbellatus flowering rush
Brassicaceae Crambe maritima seakale
Dioscoreaceae Dioscorea communis black bryony
Fabaceae Pisum sativa common pea
Fabaceae Vicia sativa common vetch
Fabaceae V. sepium bush vetch
Fabaceae V. cracca tufted vetch
Fabaceae Lathyrus sylvestris everlasting pea
Fabaceae Lathyrus latifolius bitter pea
Fabaceae Lathyrus sativus grass pea
Fabaceae Lathyrus ochrus cyprus pea
Fabaceae Lathyrus cicera red pea
Fabaceae Lathyrus aphaca yellow pea
Fabaceae Vicia ervilia bitter vetch
Fabaceae Vicia hirsuta hairy tare
Fabaceae Vicia narbonensis purple broad vetch
Cyperaceae Cyperus longus sweet flag
Cyperaceae Cyperus esculentus tigernut

252
Cyperaceae Schoenoplectum spp. common clubrush
Corylaceae Corylus cf. avellana hazel
Liliaceae Lilium martagon turk’s cap lily
Liliaceae Erythronium dog's tooth violet
Rosaceae Potentilla anserina silverweed
Rosaceae Sanguisorba officinalis great burnet
Papaveraceae Corydalis cava corydalis
Polygonaceae Bistorta officinalis european bistort
Equisetaceae Equisetum palustre marsh horsetail
Menyanthaceae Menyanthes trifoliata bogbean
Typhaceae Typha latifolia reedmace
Poaceae Avena elatior false oat-grass
Poaceae Avena sativa common oats
Poaceae Brachypodium pinnatum false brome
Poaceae Festuca sp. fescue
Poaceae Deschampsia cespitosa hair grass
Poaceae Echinochloa crus -galli barnyard grass
Poaceae Dactylis glomerata cocksfoot grass
Poaceae Elymus repens couchgrass
Poaceae Hordeum murinum wall barley
Poaceae Hordeum bulbosum bulbous barley
Fagaceae Castanea sativa sweet chestnut
Fagaceae Quercus ilex subsp. rotundifolia holm oak
Fagaceae Quercus coccifera kemes oak
Fagaceae Quercus faginea portugese oak
Smilacaceae Smilax aspera rough bindweed
‎Dennstaedtiaceae Pteridium sp. bracken
Ranunculaceae Ficaria verna lesser celandine
Nymphaea Nuphar lutea yellow waterlily
Nymphaea Nymphaea alba white waterlily
Trapaceae Trapa natans water caltrop

253
254
 Summary

Dietary studies have transformed our understanding of hominin

palaeobiology. Establishing the diet of Neanderthals has been crucial for
understanding the basis of their ecology. The gradual expansion of research using
macromammal remains has allowed substantial insights into Neanderthal meat
consumption, hunting techniques, and social cooperation. More recent approaches in
archaeological sciences such as dental wear, isotope, biomarker, and dental calculus
analyses have allowed considerable strides to be made in building a more complete
view of their dietary ecology, as they have raised further questions.

Zooarchaeological and isotopic research has shown that Neanderthals widely

hunted large- and medium-size game. Further evidence of the importance of animal-
based foods is also apparent from analysis of wear on teeth, which suggest that
across their range Neanderthals relied heavily on animal foods. Recent foragers from
northern environments, purportedly analogous to those in which some Neanderthal
populations lived, have shown that staples other than animal foods may provide up
to half of total dietary energy. However, researchers still debate the role of non-
animal foods in Neanderthal subsistence. Palaeodiet studies have so far allowed
only partial recovery of the Neanderthal diet, as there are biases against the survival
of many foods particularly plants. An alternative means of exploring plant use is to
retrieve microbotanical particles that become trapped in dental plaque as it calcifies
into hard dental calculus. Some of this debris can be identified to specific plant taxa
and plant parts. This methodology can provide direct evidence of foods and other
substances that have entered the mouth. Yet it is still unclear whether the data
generated does represent diet and other behavioural activities, because it could
possibly be merely a product of background plant debris, which is unrelated to diet.
To test the impact of these potential problems so that a working methodology could
be applied to Neanderthals, my project assessed the reliability of dental calculus in
reconstructing diets of individuals in two different populations. The first is an
archaeological sample from southeastern Iberia, and the second is a skeletal
population of West African chimpanzees. Using a comparative analysis, my project
was able to identify the utility of dental calculus to shed light not only on diet, but
also on other life history traits. The results from the Iberian assemblage prompted
me to question whether dietary reconstructions based on traditional dental calculus
studies can stand alone. Microremains from these populations exhibited evidence for
behaviours such as water intake and inhalation of airborne suspensions, as well as

255
diet. This work also characterised some of the microenvironments that likely shield
plant microremains from degradation. In the next stage of analysis, I assessed how
accurately dental calculus can record long-term diet by examining plant
microremains preserved in chimpanzees from a population with over two decades
of documented dietary history. The project then built an extensive reference
collection of Côte d'Ivoire forest plants for microremain identification, and compared
the calculus microremains to the expected presence of plants based on examination
of the feeding records. The results indicate that microremains accumulate as long-
lived dietary markers and some can reflect the proportions of plants in the diet.

After consolidating dietary inferences obtained from the study of plant

microremains in dental calculus, my dissertation analysed Neanderthal dental
calculus from five archaeological sites to document variation in Neanderthal plant
consumption. This identified reliable evidence of plant use across different regions.
Observed remains were positively identified as Neanderthal foods by comparing
plant microremains retrieved from calculus to a control dataset. We then used
published literature and databanks with a substantial database of Eurasian plants to
identify food plants across Neanderthal range. Contrary to expectations, when the
results were compared with data from previous studies, we found the breadth of
food plants with evidence of consumption was unrelated to Eurasian eco-geography.
This suggests that although southern Neanderthals relied on plants considerably
more than northern populations, Neanderthals in central and northern Europe still
consumed a diversity of plants that are comparable with Neanderthals from
southern territories. These findings open new perspectives on Neanderthal dietary
ecology. Overall, Neanderthals were foragers that combined plant use with large-
scale mammal hunting as part of a unique dietary niche adapted to Eurasia.

256
 Samenvatting

Dieetstudies hebben ons begrip van de biologie van vroege mensachtigen

getransformeerd. Vaststelling van het dieet van de Neanderthalers is van cruciaal
belang geweest voor het begrijpen van hun ecologie. De geleidelijke ontwikkeling
van onderzoek aan zoogdierresten heeft substantiële inzichten geleverd in de
vleesconsumptie, jachttechnieken en sociale samenwerking van Neanderthalers.
Hoewel recente benaderingen in de archeologische wetenschappen zoals
tandslijtage, isotopen-, biomarker- en tandsteen-onderzoek hebben geleid tot
aanzienlijke vooruitgang bij het verkrijgen van een meer compleet beeld van hun
dieetecologie, hebben ze ook nieuwe vragen opgeworpen.

Zooarcheologisch en isotopenonderzoek heeft aangetoond dat

Neanderthalers wijdverbreid jaagden op groot en middelgroot wild. Verder bewijs
voor het belang van dierlijke voedingsmiddelen blijkt uit de analyse van
tandslijtage, dat suggereert dat Neanderthalers over hun gehele verspreidingsgebied
zeer afhankelijk van dierlijk voedsel waren. Recente jager-verzamelaars uit
noordelijke gebieden, die onder ogenschijnlijk analoge omstandigheden leefden als
sommige Neanderthalerpopulaties, hebben laten zien dat tot maximaal de helft van
hun totale energie opname afkomstig is van niet-dierlijk voedsel. Desondanks
bediscussiëren onderzoekers de rol van niet-dierlijk voedsel in het levensonderhoud
van Neanderthalers nog altijd. Palaeodiëtaire studies hebben tot nu toe een
incompleet beeld gegeven van het dieet van Neanderthalers vanwege een negatief
effect op de preservering van sommige voedselresten, in het bijzonder dat van
plantaardig voedsel.

Een alternatieve methode om het gebruik van planten te onderzoeken is door

het bestuderen van microbotanische deeltjes die vast komen te zitten in tandplak als
dit calcificeert tot tandsteen. Een deel daarvan kan geïdentificeerd worden tot
specifieke plantentaxa en plantendelen. Deze methode verschaft direct bewijs voor
voedsel en andere substanties die de mond binnen gekomen zijn. Echter, het is nog
altijd onduidelijk of de data dieet en andere gedragsactiviteiten vertegenwoordigen
omdat ze ook het gevolg kunnen zijn van plantendelen die ongerelateerd zijn aan
dieet. Om de impact van deze potentiële problemen te testen, zodat een werkende
methode kan worden toegepast op de Neanderthalers, onderzocht ik in mijn project
de betrouwbaarheid van tandsteen in het reconstrueren van diëten van individuen
in twee verschillende populaties. De eerste is een archeologische populatie uit
Zuidoost-Iberië, en de tweede een populatie van West-Afrikaanse chimpanzees.
Door vergelijkend onderzoek in mijn projects was het mogelijk het nut van
tandsteenonderzoek voor diëtaire studies aan te tonen, maar ook met betrekking tot
andere levensgeschiedeniskenmerken. De resultaten van de Iberische populatie

257
leidden me ertoe me af te vragen of dieetreconstructies gebaseerd op traditioneel
tandsteenonderzoek op zichzelf kunnen staan. Microresten van deze populaties
vertegenwoordigen bewijs voor de consumptie van water en de inhalatie van
luchtdeeltjes , naast dieet. Verder karakteriseert dit onderzoek enkele micro-
omgevingen die waarschijnlijk plantaardige microresten beschermden tegen
degradatie. In de volgende fase onderzocht ik hoe nauwkeurig tandsteen het dieet
kan reflecteren op de lange termijn door microresten te onderzoeken bij chimpansees
uit een populatie met meer dan twee decennia aan gedocumenteerde
dieetgeschiedenis. Vervolgens verzamelde het project een uitgebreide
referentiecollectie van bosplanten uit Ivoorkust voor de identificatie van
microresten, en vergeleek de geobserveerde microresten uit tandsteen met de
verwachtte aanwezigheid van planten op basis van de dieetgeschiedenis. De
resultaten geven aan dat microresten uit tandsteen accumuleren als dieetindicatoren
over de lange termijn en dat sommigen het aandeel van planten in het dieet
reflecteren.

Na het bevestigen van dieetreconstructies op basis van het onderzoek aan

microresten uit tandsteen, analyseerde ik Neanderthalertandsteen van vijf
archeologische vindplaatsen om variatie in het gebruik van plantaardig voedsel in
Neanderthalers te documenteren. Dit resulteerde in betrouwbaar bewijs voor het
gebruik van planten in verschillende gebieden. Waargenomen overblijfselen werden
positief geïdentificeerd als Neanderthaler voedsel door het vergelijken van de
plantenmicroresten uit tandplak met een controledataset. Vervolgens gebruikten we
gepubliceerde literatuur en databases die een substantieel deel van de Euraziatische
planten bevatten om plantaardig voedsel te identificeren voor het gehele
verspreidingsgebied van de Neanderthalers. In tegenstelling tot onze verwachingen
vinden we, als we onze resultaten combineren met data van voorgaande studies, dat
de diversiteit van plantaardig voedsel met bewijs voor consumptie ongerelateerd is
aan de Euraziatische ecogeografie. Dit suggereert dat hoewel zuidelijke
Neanderthalers meer afhankelijk waren van plantaardig voedsel dan noordelijke
populaties, Neanderthalers in Centraal en Noord-Europa wel een diversiteit aan
planten consumeerden die vergelijkbaar was met die van Neanderthalers in
zuidelijke gebieden. Deze resultaten geven een nieuw perspectief op de
dieetecologie van Neanderthalers. In het algemeen waren Neanderthalers jager-
verzamelaars die het gebruik van planten combineerden met het jagen op groot wild
als onderdeel van een unieke niche aangepast aan Eurazië.

258
 Acknowledgements

I am deeply grateful to my supervisor and group leader Amanda Henry.

During my studies in the group, Amanda Henry supported my work and ensured
my development, patiently to a fault. In no way could the project have advanced as
smoothly without her foresight, expertise and generosity, even letting me use lab
resources after my dissertation submission. I am in appreciate not just for giving me
the opportunity to write a doctorate in the Plant Group but for sharing four years of
formation. I am grateful to second supervisor, Domingo C. Salazar-García for who
guided me through the PhD process. He always selflessly gave time to ensure every
spontaneous deadline could be met. His contributions to my doctoral thesis and my
formation are not easily summarised or quantified. Jean-Jacques Hublin deserves
thanks for his insightful comments through my studies. He always made me feel
welcome as a Research Group member hosted by the Department of Human
Evolution. Wil Roebroeks, for his tips and mentoring along the way.

Aside from my supervisory team, many others contributed to the progress of

the PhD. Fellow PhD candidate Stephanie who deeply enriched my perspectives on
the evolution of nutrition. Chelsea Leonard on many occasions kindly gave her time
to critique and develop my ideas and papers. Shira Gur-Arieh, for her grounding
comments and pharmacological support.

This project was carried forward immeasurably by the support of the

Department of Primatology and the Taï Forest Project staff. Roman Wittig’s
openness to arrange the collection of plants in the Taï Forest made the convoluted
task of building a reference collection possible. I am indebted to Roger Kami Nabo
and the other staff in the forest who took the time to collect the reference plants, dry
fresh plants and return them to Leipzig. I also wish to note the support of the
Ministère de la Recherche Scientifique and the Ministère de l’Environnement et des
Eaux et Forêts of Côte d’Ivoire, Centre Suisse de Recherches Scientifiques at Abidjan,
Côte d’Ivoire, the Office Ivorien des Parcs et Reserves, the Director of the Taï
National Park for authorisation of this research. Utah Schwarz sorted out all the
practicalities of sampling the chimpanzee skeletal collection. Julia Riedel answered
many questions on working with Taï Chimpanzee observational data. Loki du Toit
made it possible for me to use these observation records. Geraldine Fahy, Gottfried
Hohmann and Tobias Deschner took the time to fill me in with much essential

259
background information and providing valuable plant samples. I am also thankful to
the Head of Primatology Christophe Boesch who kindly allowed my sampling of the
Taï Skeletal Collection.

Many helpful archaeological scientists assisted the project by advising on

microremain and material identifications such as Marco Madella, Paschal Verdin,
Sanjay Eksambekar, Arlene M. Rosen, Irwin Rovner Linda Perry, Angela Perri, Vera
Aldeias, and Dan Cabanes. The extensive statistics and programming support from
Colleen Stephens and Roger Mundry without whom I would not have complete the
projects data analysis. Maria Bendžuchová helped by providing spectroscopy
technical advice. I am also grateful to Philip Gunz, Layne Vashro, and Andre Strauss
who also helped on data analysis and programming.

I am keen to mention Ian Reid of Nano Imaging and Material Analysis Centre
at the University College Dublin for SEM-EDX analysis of many dental calculus
samples. Pat O’Reilly of Nature First, genially identified spores that were found in
certain dental calculus samples. A great number of people provided the
archaeological contexts of the many sites where I sourced dental calculus.
Authorities such as Bence Viola, Ivor Karavanić, Julià Maroto Genover, Philip Nigst
and Fred Smith kindly gave the time to give all the background information the
project required.

The diligence, professionalism and kindness of the group’s technical staff

Antje Hutschenreuther, Joerg Watzke, Simone Schmidt, Thomas Büdel, Annabell
Reiner and Steffi Hess deserve a lot of credit. Their work unfailingly kept the labs
and protocols running when other events took my attention. For the time spent
outside the laboratories, I thank Victor Weiler, Silke Streiber, and Conny Schicke for
administrative assistance and helping me adapt to Germany and Lukas Westphal for
superb ICT support. Also helpful were, Roswitha Manning and Anneke Hendriks
for Leiden University assistance. Marie Soressi, for support at the Faculty of
Archaeology in Leiden University. Frido Welker, for translating my doctorate
summary into Dutch and Karen Ruebens for QGIS advice.

I appreciate the archaeologists and curators who saw the value of this project
and provided access to skeletal material for sampling. Michael J. Walker (University
of Murcia, Spain) permitted access to the Sima de las Palomas del Cabezo Gordo
material, and Mariano López helped to make our collaboration happen. Jadranka
Lenardić, Dejana Brajkovic and Ivan Gusic (Croatian Academy of the Sciences,
Croatia) allowed me to sample the Vindija hominin remains. Rosa Alsius (Farmàcia

260
Rosa Alisus in Banyolas, Spain) allowed me to access to the Banyolas mandible.
Raffaele Sardell and Mauro Rubini (Servizio di Antropologia, Italy) permitted access
to Grotta Guattari and Grotta Fossellone. Andreas Darlas (Greek Ministry of
Culture) and Katerina Harvati-Papatheodorou (Eberhard-Karls-Universität
Tübingen, Germany) permitted access to the Kalamakia remains. Joaquín Lomba,
María Haber and Azucena Avilés for kindly allowed access to the Camino del
Molino calculus samples. In addition, the project benefited from plant samples
provided by Martin Freiberg from the Leipzig Botanical Gardens. The doctorate
required drawing approaches and data from many different disciplines outside my
training such as palaeoclimatology to take into account all the factors that affected
Neanderthals. It is due to the advice of Dave Pollard, Robert Foley, Kathryn
Fitzsimmons and William Davies that integrating palaeoclimatology approaches was
successful.

A word of acknowledgement is needed for my friends and fellow Max Planck

students and colleagues; Nicole Antonette del Rosario for helping me develop my
thoughts thoroughly and for diligently proofreading my work, Cynthianne Debono
Spiteri, Mariska Carvalho, Nadia Scott, Sylvio Tüpke, Liu Huan, Tao Chen, Marcello
Mannino, Zewdi Tsegai, Collin Moore, Nina Doerschner, Dawit Desta, Nick
Stephens, Viktoria Oelze, Stefanie Stelzer, Will Archer, Adeline Le Cabec, Sahra
Talamo, and Vaidas Suncovas at the Max Planck Institute for Evolutionary
Anthropology, who raised my spirits during the hard slog of the doctorate. Each
made the institute a uniquely enriching place to work.

My family played a large role in writing this dissertation. I am appreciative to

my brother John for proof reading many drafts along the way and Eunhee for her
kindness. At this point I must acknowledge my parents John and Julie Power, whose
endless encouragement above all motivated me. Their personal sacrifices for me not
only made it all happen but have been an inspiration.

261
262
 Contributions
Robert C. Power has designed the research with input from Dr Amanda G. Henry
and Dr Domingo C. Salazar-García, done all lab work, data analysis, written the
papers and interpreted results. Dr Amanda G. Henry and Dr Domingo C. Salazar-
García have supervised this PhD. Both gave feedback and guidance along all steps of
the process to Robert C. Power.

Contributions to each paper

Paper 1:

Assessing use and suitability of scanning electron microscopy in the

analysis of micro remains in dental calculus

This paper was written following analysis of the dental calculus from two
populations with multiple means of microscopy. As author, I developed the concept
of comparing microremain data gathered from different methods of microscopy
(optical microscopy/scanning electron microscopy and energy-dispersive X-ray
spectroscopy). It allowed me to develop an approach to assess the potential of
energy-dispersive X-ray spectroscopy to detect starch exposed to salivary amylase. I
performed all sampling of dental calculus and all microscopy and spectroscopy, but
scanning electron microscopy and spectroscopy was conducted with assistance of
Ian Reid for SEM–EDX analysis (NIMAC UCD). Roman Wittig and Domingo C.
Salazar-García provided materials. I prepared and wrote the manuscript, with input
from all authors and guidance of my supervisors (A.G.H and D.C.S.G.).

Paper 2:

Dental calculus evidence of Taï Forest Chimpanzee plant consumption and

life history transitions

This project was conducted in four steps- initially I employed Taï

observational records to develop a chimpanzee diet profile. This was used to build
an extensive reference collection of the microremains present in Taï Chimpanzee
diet, allowing me to create a replicable method to classify microremains frequent in
their diet. Lastly, I analysed chimpanzee dental calculus and matched identified
starch and phytoliths to dietary records. I sampled calculus and conducted
microscopy of all dental calculus and plant reference material. Roman Wittig

263
advised on using Taï Forest data to meet research goals, and organised the collection
of plant reference samples from the Taï Forest National Park. This reference
collection was supplemented with plants from the Institute of Botany that were
supplied by Martin Freiberg. Geraldine Fahy supplied some useful additional plant
reference samples. I conceptualised the statistical approach to identifying
microremains found in dental calculus with input from Amanda G. Henry, Colleen
Stephens and Roger Mundry, these last two helped me to design the statistical
analysis methodology. I performed all of the analysis required for the project. This
was completed in a framework devised in with liaison with Amanda G. Henry,
Roman Wittig and Domingo C. Salazar-García. I compiled the findings, wrote the
paper, and edited it along with all authors. Amanda G. Henry and Domingo C.
Salazar-García, supervised the process.

Paper 3:

Dental calculus indicates widespread plant use within the Neanderthal

dietary niche

My contribution to this paper involved designing and leading the project at

all stages with support of Amanda G. Henry and Domingo C. Salazar-García. This
study of Neanderthal dental calculus required traveling for sampling to a variety of
museums and research institutions that curated Neanderthal and fauna remains. The
Kalamakia teeth were sampled in Leipzig by Amanda G. Henry, and the Sima de las
Palomas del Cabezo Gordo samples were taken in the University of Murcia by
Amanda G. Henry and Domingo C. Salazar-García. I examined each sample of
dental calculus using microscopy. I identified all finds with plant reference material
and prepared the results. Drs. Domingo C. Salazar-García, Mauro Rubini and
Jadranka Lenardic assisted with interpreting archaeological site data. I then
accompanied pollen, charcoal and fauna with palaeotemperature simulations from
the Stage Three Project, and used it to build a model of dietary breadth with the
guidance of Colleen Stephens and Roger Mundry. I wrote the article with the
guidance of Amanda G. Henry, Domingo C. Salazar-García. Amanda G. Henry and
Domingo C. Salazar-García helped me revise and finalised the article.

264
 Curriculum vitae

The author of this dissertation, Robert C. F. Power, was born in Waterford in

Ireland. After receiving his Leaving Certificate at Waterpark College in 2007, he read
for a Single Honours Bachelor in Arts (Archaeology) in University College Cork,
National University of Ireland. In Cork he began his interest in ancient diets by
studying macro and microbotanical research. During this time participated in
archaeological excavations of prehistoric and historic sites in Ireland and Canada. In
addition, he has carried out post-excavation analysis as an archaeobotanical research
assistant at the Discovery Programme in Dublin, Ireland. He was also an officer in
the University’s Medieval and Renaissance Society. His interest in the complexity of
human interactions with the natural world led him to specialise in Environmental
Archaeology. He pursued this in 2010 with a Master of Science in Environmental
Archaeology from the Institute of Archaeology, University College London in the
University of London. He received his Master of Science in 2011 with a dissertation
on using phytoliths from Raqefet Cave in Israel to infer plant use by Natufian
foragers. After the Master of Science, he started his PhD in 2011, focusing on dental
calculus in dietary studies and reconstructing Neanderthal plant use in the Plant
Foods in Hominin Dietary Ecology Research Group at the Max Planck Institute for
Evolutionary Anthropology. In addition to the doctorate research, he has presented
his work at international conferences and published his work in several peer-
reviewed, international journals such as the Journal of Anthropological Archaeology,
Quaternary International and the Proceedings of the National Academy of Sciences.
He is now employed as a research fellow at the Max Planck Institute for the Science
of Human History in Jena in Germany.

265